System for the interpretation and querying of content generated by an artificial intelligence model

Abstract

A system for querying interpreted content from a generative artificial intelligence model is disclosed. The system includes a generative artificial intelligence model and one or more processors. The model is configured to accept a natural language text input and produce a human-like text response. The system performs operations including receiving a natural language text input representative of a query subject and a query context, receiving an input representative of a query response structure, amplifying a combination of the inputs, providing the amplified query to the model, receiving the human-like text response generated by the model, accumulating the responses to form a global query response, and providing the global query response to a computerized interface.

Description

[0001] TITLE: SYSTEM FOR THE INTERPRETATION AND QUERYING OF CONTENT GENERATED BY AN ARTIFICIAL INTELLIGENCE MODEL

[0002] BACKGROUND OF THE INVENTION

[0003] The present disclosure pertains to the expansive and rapidly evolving field of artificial intelligence (Al), a discipline that encompasses a wide array of technical areas including machine learning, natural language processing, and deep learning. Within this broad field, the disclosure narrows its focus to the specialized sub-field of generative artificial intelligence models. These models, often powered by deep learning algorithms, are designed to generate new content that is similar to the input data they have been trained on.

[0004] Generative artificial intelligence models have found applications in a variety of sectors, from creating realistic images and synthesizing human-like text to generating music and even designing pharmaceutical drugs. In particular, the use of generative artificial intelligence models for natural language processing tasks has seen a surge in interest.

[0005] Such applications also refer to image or video generation. Basically, the core idea is to transform words into content, which leads to words being the focal point of such content. It is however not intuitive to ask, for example, for the generation of an image by defining lens obturation, iso parameters, white balance, view angle or colorimetric space.

[0006] These tasks often involve the generation of human-like text responses based on a given input, a capability that has profound implications for industries such as customer service, content creation, and digital marketing. Such implications affect any task assisted by a computer assuming it requires reflection, processing, input, editing, classification or analysis.

[0007] One common application of generative artificial intelligence models in the realm of natural language processing is the creation of chatbots or virtual assistants. These Al- powered entities are capable of understanding and responding to natural language queries, providing users with a more intuitive and human-like interaction experience. However, the quality and relevance of the responses generated by these models can vary greatly depending on parameters that are external to the query as such.

[0008] In particular, the quality of the response to the query depends on "model fatigue", which corresponds basically to the amount of computing power available at the time of sending the query, which results in irregular and variable responses to a singular input.

[0009] The quality of the response also depends on “model intelligence”, which is the capacity of a model to effectively consider the context of a request.

[0010] This inconsistency can lead to a less than satisfactory user experience, as the same query can yield different responses depending on the available computing power at the time of query processing. This variability can undermine the reliability of the artificial intelligence model, making it less effective for applications that require consistent and reliable responses, such as customer service or digital marketing.

[0011] Furthermore, the precision and length of the content generated by the artificial intelligence model are influenced by the quality and length of the initial query. This means that the input query plays a substantial role in determining the output. A well-structured, detailed, and lengthy query is more likely to yield a precise and comprehensive response. But if a query is too long, context saturation induces objective conflicts and attention selection biases. Conversely, a short or vague query may result in a less precise and shorter response. This dependency can pose challenges, especially when the initial query lacks sufficient detail or length, systematically leading to less accurate or incomplete responses. This, in turn, requires the development of specific skills for users, known generally as "prompt engineering". However, such skills typically amount to query reformulation.

[0012] The following should be taken into consideration to describe the current state of technology. What can be called the unconscious of words consists of everything that each word and phrase evoke in the mind of the speaker.

[0013] Although words have definitions that serve as reference points, their interpretation is as varied as it is unpredictable.

[0014] This is easily understood when we use adjectives, as everyone can recognize differences in the perception of beauty, for example. The adjective "beautiful" is subjective and may evoke the beauty of a landscape for one person or the beauty of the ocean for another.

[0015] These unspoken elements that travel through instructions and requests characterize the unconscious of words.

[0016] It is within this unconscious material of words that the explanation for generalization in the attention-processing mechanism of a Transformer model lies.

[0017] Indeed, every word used by a human invokes billions of possible interpretations for the recipient of these words. When artificial intelligence is this recipient, having simultaneous access to the entirety of available intellectual material, it must arbitrate between billions of interpretative possibilities to generate its response.

[0018] However, without knowing or understanding the frame of reference of its interlocutor, this recipient — artificial intelligence — cannot, in any way, produce a response that perfectly aligns with the speaker's intentions, as those intentions carry an unconscious dimension conveyed through unspoken words. This should be a prerequisite for any request whose result could be explained. It is therefore normal to be unable to explain the black box’s processing mechanism, given the vast amount of information involved and the fact that the logic of association and distance between different corporal and vectorial schemas is not based on hierarchical reference points accessible to humans.

[0019] Labeling and categorizing data are insufficient in this regard.

[0020] This is because the relationships that the transformation process establishes to produce a result remain inaccessible to the human mind, which needs reference points to understand.

[0021] These reference points exist; it is simply a matter of recognizing their true value. Summary of Invention

[0022] The present invention aims to solve at least part of the technical problems listed above.

[0023] TO BE ADDED

[0024] As such, according to a first aspect, the present invention aims at a system for querying interpreted content from a generative artificial intelligence model, comprising:

[0025] - a generative artificial intelligence model based on an attention mechanism configured to accept an input in the form of a natural language text and configured to use deep learning to produce a human-like text response as a function of the accepted natural language text input; and

[0026] - one or more processors; and

[0027] - one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the computing system to perform operations, the instructions being representative of steps of:

[0028] - defining a query response framework, comprising:

[0029] - receiving a natural language text input representative of a query context,

[0030] - receiving a hierarchical natural language text input representative of at least one implicit and hierarchically defined user-related query context,

[0031] - receiving an input representative of a query response structure, said structure comprising at least two individual segment type identifiers,

[0032] - setting at least one preset query generation rule, in the form of natural language to form a library of preset rules, each said preset query generation rule being associated with at least one segment type identifier, - receiving a natural language text input representative of a query subject,

[0033] - for each segment type identifier:

[0034] - forming a combination of the inputs representative of the query response framework and query subject,

[0035] - providing the combination to the generative artificial intelligence model,

[0036] - receiving the human-like text response generated by the generative artificial intelligence model as a function of the provided combination,

[0037] - accumulating the human-like text responses received for each segment to form a global query response, and

[0038] - providing the global query response to a computerized interface.

[0039] Thanks to these provisions, firstly, the system allows for more precise, comprehensive and better aligned responses from the artificial intelligence model by amplifying the input query with preset rules and induced interpretation modalities by concatenation and / or accumulation of frameworks of segment types. This can lead to improved user experience and more effective use of artificial intelligence in applications such as customer service, digital marketing, financial analysis, research and development, or medical anamnesis. Secondly, the system addresses the issue of "model fatigue" by structuring the query in a way that reduces the dependency on the available computing power at the time of query processing by possibly desaturating context to focus the objective on a given direction via the expected response structure. This can lead to more consistent and reliable responses from the artificial intelligence model. Thirdly, the system allows for the accumulation of responses to form a global query response, which can provide a more complete and detailed answer to the user's query. Fourthly, the use of preset query generation rules allows, for each segment, to obtain a different response from the artificial intelligence model targeting the subject of the query according to different conceptual query approaches, each query approach corresponding to a segment. For example, a query approach may correspond to a historical perspective on a subject. Another query approach may correspond to a legal perspective on a subject. By forcing the artificial intelligence model to provide two separate responses for each perspective or approach, the quality of the response is improved.

[0040] In particular embodiments, during the step of receiving an input representative of a query response structure, said structure comprises at least one sub-segment type identifier, said sub-segment type identifier being associated with a segment type identifier or with a sub-segment type identifier. Such embodiments allow for a more granular and structured approach to query processing, which can enhance the precision and relevance of the artificial intelligence model's responses. By associating each sub-segment type identifier with a segment type identifier or another sub-segment type identifier, the system can better understand and interpret the context and intent of the query. This can lead to more accurate and contextually appropriate responses. Furthermore, this structure can help mitigate the effects of "model fatigue" by distributing the processing load across multiple sub-segments, systematically leading to more consistent and reliable responses. If each segment is subject to an individual request, associated with segment structure information for the parent segment in order to maintain coherency, each said segment may then be provided to different generative models, thus increasing the relevance of the output.

[0041] In particular embodiments, the system object of the present invention further comprises:

[0042] - prior to the step of receiving a natural language text input representative of a query subject, a step of providing, to a computerized interface, a preset natural language text input comprising an incomplete natural language text,

[0043] - a step of completing, by a user upon the computerized interface, the preset natural language text input by adding natural language text to the preset natural language text, said completed preset natural language text being used during the step of receiving a natural language text input representative of a query subject.

[0044] Such embodiments increase the interpretation capability of the request.

[0045] Such embodiments provide the following benefits: firstly, it guides the user in formulating a query, which can lead to more effective queries and better responses from the artificial intelligence model. Secondly, it can help ensure that the query is in a format that the artificial intelligence model can process effectively, which can improve the quality of the artificial intelligence model's responses. Thirdly, it can make the system easier to use, especially for users who are not familiar with formulating queries for artificial intelligence models. This is even more true if anthropomorphous anchors are proposed to a user and known from the generative model, such as a job title, first name or skills, as anthropomorphous contextualization participates in leading to formulations which imply the subconscious of words aligned with the object of the provided anthropomorphous anchors.

[0046] In particular embodiments, at least one preset generation rules is defined by a hybrid query input, comprising both natural language text input and interpretable programming language text input, said hybrid query input being used during the step of forming a combination. In such embodiments, the use of a hybrid query input allows the system to leverage the precision and expressiveness of programming languages to better specify the query, which can further enhance the quality of the artificial intelligence model's responses. This includes, in particular, creating routines of instructions to be interpreted, and not executed, by the artificial intelligence model.

[0047] The hybridization of natural language and interpretable programming language allows:

[0048] - the interpretation of natural language by convention, which generates less subjective interpretation, which leads to better results,

[0049] - the destructuration of statistically identical probability-based results by introducing elements of randomness,

[0050] - the definition of the quantity of work to be performed by the generative artificial intelligence which leads to better respect for the instructions given, and

[0051] - the reduction in the length of the instructions given.

[0052] In particular embodiments, the step of interpreting comprises:

[0053] - a step of dynamically instantiating interpretable programming language variables representative of natural language text elements as a function of at least one of:

[0054] - the received natural language text input representative of a query subject,

[0055] - the received natural language text input representative of a query context, and

[0056] - the received input representative of a query structure,

[0057] - a step of transposing, in the hybrid query input, said natural language text elements by said instantiated variables in the interpretable programming language to form an adapted hybrid query input, said adapted hybrid query input being used during the step of forming a combination.

[0058] Such embodiments allow for a more nuanced and contextually aware processing of the query, potentially leading to more accurate and relevant responses from the artificial intelligence model. Furthermore, by incorporating programming language variables into the query, the system can leverage the precision and expressiveness of programming languages to better specify the query, which can further enhance the quality of the artificial intelligence model's responses.

[0059] In particular embodiments, at least one generation rule is configured to generate interpretable programming language text representative of an execution iteration as the function of a programming language variables representative of natural language text element. Such embodiments allow for a more nuanced and contextually aware processing of the query, potentially leading to more accurate and relevant responses from the artificial intelligence model. Furthermore, by incorporating programming language variables into the query, the system can leverage the precision and expressiveness of programming languages to better specify the query, which can further enhance the quality of the artificial intelligence model's responses.

[0060] In particular embodiments, the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query intention.

[0061] In particular embodiments, the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query objective.

[0062] In particular embodiments, the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query audience.

[0063] In particular embodiments, the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query audience culture. Such a culture can correspond to a set of descriptor keywords of said culture.

[0064] In particular embodiments, the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query skillset.

[0065] In particular embodiments, at least one preset generation rule is set as a function of the query skillset received.

[0066] In particular embodiments, the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query objective.

[0067] Such embodiments allow for a more detailed and contextually relevant input for the artificial intelligence model, which can lead to more accurate and appropriate responses. It also provides a structured way to capture and represent various aspects of the query context, enhancing the system's ability to understand and respond to the user's query effectively.

[0068] In particular embodiments, at least one implicit and hierarchically defined user- related query context input is representative of:

[0069] - a cultural context applicable to the user,

[0070] - a religion of the user,

[0071] - a moral philosophy followed by the user, and

[0072] - an ideology of the user.

[0073] Such embodiments allow for the obtaining of contextually relevant responses from the generative artificial intelligence model, wherein the context is not explicitly set by a user and defined by the corpus associated with the hierarchically defined user-related query context input.

[0074] In particular embodiments, at least one implicit and hierarchically defined user- related query context input is associated with a hierarchical rank value.

[0075] In particular embodiments, during the step of forming, the hierarchical natural language text input is used as a prefix to the combination, the hierarchical natural language being sequentially added to the combination, by order of decreasing hierarchical rank value.

[0076] In particular embodiments, during the step of forming, a digital identifier representative of an intention perspective is included in a request, the variations of perspective allowing the processing of the following requests:

[0077] - requests whose intentions are shared both by the intervention framework (referred to as the project) and by interlocutors outside this framework, invoking unified contexts for both public-facing and private internal use,

[0078] - requests whose intentions can only originate from users actively participating in the execution framework, as members with specific roles, professions, or assigned tasks, invoking contexts limited to the connected user, always within the reference framework, and

[0079] - requests whose intentions can only come from disconnected users not belonging to the intervention framework but interacting with it, invoking associated interpretative contexts while adhering to the orientations of the solicited environment’s reference framework.

[0080] "Personal" contexts are only propagated to the current user. Thus, user-specific elements are known to all generative systems they interact with, without affecting the initial parameters of those systems. Each generative system receives distinct parameter elements depending on the user consulting it, enabling the necessary variability to adjust intentions based on explicit preferences.

[0081] "General" contexts refer to information and / or skills globally imposed on all users (whether connected or not) acting within the dedicated environment. These contexts, regardless of the query’s perspective, always remain within the bounds of the reference framework of global interpretative values. General contexts are systematically considered.

[0082] An anthropomorphic agent can be defined by a profile:

[0083] - a name, or an anthropomorphic marker that introduces cultural and / or temporal notions, as names and surnames often carry cultural connotations and suggest an era (A marker for both humans and generative systems), - an age, or an anthropomorphic marker used to invoke and / or emphasize an expression style or ideological stance generally associated with age (A marker for both humans and generative systems),

[0084] - a gender, or an anthropomorphic marker used to invoke and / or emphasize a gendered perspective and align expression accordingly. (A marker for both humans and generative systems),

[0085] - a profession, or an anthropomorphic marker used to reference disciplinary aspects (corpus) related to a profession from a human perspective,

[0086] - a primary function, which reinforces occupational focus on a specific objective while improving the interpretation of imperative instructions.

[0087] - a description, which helps strengthen the generative system’s position relative to its interlocutor, clarifying its role in the conversation and ensuring better alignment,

[0088] - a workplace, which delimits the interpretative scope to a particular environment,

[0089] - a language level, which directs sentence construction and morphology,

[0090] - an expression level, which enriches sentence construction for more relevant and polished communication,

[0091] - a style, which maintains a stylistic expression throughout discussions, beyond merely labeling the style - this enables the invocation of literary devices such as metaphors, figures of speech, alliteration, hyperboles, and / or

[0092] - a register, which maintains a consistent register throughout discussions, specifying its nature - this supports the invocation of specific jargon, lexicons, or thesauri.

[0093] A query frame may comprise:

[0094] - a fill-in-the-blank prompt, which predetermines the query’s direction for optimal alignment with the generative system’s capabilities, skills, and anthropomorphic characteristics, and / or

[0095] - a discussion starter, which sets the tone and objectives for the conversation - for example, "How can I help you?" invites broader possibilities compared to "I am ready to analyze your chart!"

[0096] A query scope may comprise:

[0097] - an intervention context, which defines the scope for applying skills, knowledge, and anthropomorphic characteristics, enhancing jargon and focus while aligning with objectives,

[0098] - an audience type, which adjusts style and register based on the recipient’s profile,

[0099] - a set of objectives or tasks, which explicitly command primary goals around which all contextual characteristics are articulated, - a functional role, which reinforces the understanding of primary objectives and arbitrates conflicts stemming from multiple competencies or conflicting instructions,

[0100] - a posture in conflicts, which specifies particular handling cases or organizational / moral policies,

[0101] - a set of skills, expressed as imperatives directed at the generative systems, reinforced by explicit objectives and conditioned by other contextual characteristics, and / or

[0102] - a set of knowledge, which is raw information serving as a corpus of priority considerations for interpretation.

[0103] An alignment criterion may be defined by:

[0104] - priority management parameters,

[0105] - “project” values inheritance, wherein the generative system adheres to the project’s interpretative values while aligning with present characteristics,

[0106] - inactive inheritance, which prioritizes current interpretative values, and / or

[0107] - interpretative values specific to generative systems.

[0108] An editorial orientation may comprise:

[0109] - an editorial angle, which corresponds to a set of imperative instructions affirming the unique perspective derived from anthropomorphic characteristics in the expression context, and / or

[0110] - an editorial line, which corresponds to a set of imperative instructions reaffirming objectives and perspectives.

[0111] In particular embodiments, the method object of the present invention comprises a step of defining a set of assistant generative systems, defined in relation to the generative system recipient of the generation query and provided to the generative system recipient of the generation query as a part of a context definition query.

[0112] Each generative system can be supported by team members with distinct anthropomorphic characteristics aligned with the primary generative system’s intentions. Team members act as subordinates under the primary generative system’s objectives, within the project’s general framework.

[0113] Secondary team members can be added for independent objectives or sequentially for processes where the primary generative system’s responses are relayed to the first team member, whose responses are then relayed to the next, and so on. This enables deep relevance for specific tasks while adhering to defined interpretative criteria and hierarchy.

[0114] In particular embodiments, an incremental expander is used and facilitates in-depth distribution by dynamically suggesting generative systems dedicated to particular tasks, enhancing focus on objectives. Such expanders regenerate new queries based on selected words, phrases, or passages, recontextualizing them within the selected content while integrating global / project- level context.

[0115] Such suggestions include:

[0116] - referential frameworks, which may be related / opposed schools of thought, mental mapping, thesauri, and / or

[0117] - a plan generation: Informative, argumentative, conversational, documentary, pamphlets, essays, satire, chronicles, reflections.

[0118] According to a second aspect, the present invention aims at a computer- implemented method for querying interpreted content from a generative artificial intelligence model, comprising the steps of:

[0119] - defining a query response framework, comprising:

[0120] - receiving a natural language text input representative of a query context,

[0121] - receiving a hierarchical natural language text input representative of at least one implicit and hierarchically defined user-related query context,

[0122] - receiving an input representative of a query response structure, said structure comprising at least two individual segment type identifiers,

[0123] - setting at least one preset query generation rule, in the form of natural language to form a library of preset rules, each said preset query generation rule being associated with at least one segment type identifier,

[0124] - receiving a natural language text input representative of a query subject,

[0125] - for each segment type identifier:

[0126] - forming a combination of the inputs representative of the query response framework and query subject,

[0127] - providing the combination to the generative artificial intelligence model,

[0128] - receiving the human-like text response generated by the generative artificial intelligence model as a function of the provided combination,

[0129] - accumulating the human-like text responses received for each segment to form a global query response, and

[0130] - providing the global query response to a computerized interface.

[0131] This second aspect provides the same benefits as the system object of the present invention.

[0132] Such an invention can be used in the context of omnidirectional communications structured by language, such as instruction reception and emission. Such an invention may impact on several technical domains, industries, and sectors. For example, and not limited to, such an invention may be used in the contexts of:

[0133] - health and well-being: therapeutic automated conversation systems require a high and contextually aligned sensibility to bring adapter and relevant care and guidance to patients,

[0134] - education: crafting tailored and dedicated content requires alignment on teaching requirements and the level of students,

[0135] - pharmaceutical and medical research: automated and reliable synthesis and analysis of medical data requires alignment on norms.

[0136] It should be noted that the present invention can be understood by analogy to the dynamic determination of appropriate phraseology for a specific query.

[0137] As a reminder, phraseology is the study of set or fixed expressions, such as idioms, phrasal verbs, and other types of multi-word lexical units (often collectively referred to as phrasemes), in which the component parts of the expression take on a meaning more specific than, or otherwise not predictable from, the sum of their meanings when used independently.

[0138] Phraseology is concerned with the way in which these expressions are formed, their meaning, their use in discourse and their role in communication.

[0139] Phraseological expressions can include idioms, proverbs, collocations, fixed expressions and cliches. They are often rooted in culture and can vary considerably from one language to another, making it difficult to translate and interpret without knowledge of the cultural context.

[0140] Whereas traditional phraseology is interested in the contextual interpretation of phrasemes by human beings, the present invention is interested in defining the appropriate interpretation of phrasemes as a function of a determined and user-defined context for a generative artificial intelligence.

[0141] BRIEF DESCRIPTION OF FIGURES

[0142] Other advantages, purposes and particular characteristics of the invention shall be apparent from the following non-exhaustive description of at least one particular embodiment or succession of steps of the present invention, in relation to the drawings annexed hereto, in which:

[0143] - Figure 1 represents a flowchart representing a particular implementation of the method object of the present invention and the corresponding instructions used in the system of the present disclosure,

[0144] - Figure 2 represents a block diagram illustrating a particular embodiment of a system object of the present invention, - Figure 3 represents an example of an incomplete natural text query to be filled by a user,

[0145] - Figure 4 represents an example of a query response structure builder to be filled in by a user, and

[0146] - Figure 5 represents a sample query frame respecting a query frame built according to the present invention.

[0147] DETAILED DESCRIPTION

[0148] This description is not exhaustive, as each feature of one embodiment may be combined with any other feature of any other embodiment in an advantageous manner.

[0149] Various inventive concepts may be embodied as one or more methods, of which an example has been provided. The acts performed as part of the method 100 may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.

[0150] The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.”

[0151] The phrase “and / or,” as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and / or” should be construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and / or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and / or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

[0152] As used herein in the specification and in the claims, “or” should be understood to have the same meaning as “and / or” as defined above. For example, when separating items in a list, “or” or “and / or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one of a number or lists of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of” or “exactly one of,” or, when used in the claims, “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e., “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.” “Consisting essentially of,” when used in the claims, shall have its ordinary meaning as used in the field of patent law.

[0153] As used herein in the specification and in the claims, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and / or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

[0154] In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of” and “consisting essentially of” shall be closed or semi-closed transitional phrases, respectively.

[0155] Figure 2 represents a block diagram that illustrates an example system 200 with which an embodiment of the method 200 and / or system 100 of the present invention may be implemented. In the example of figure 2, a computer system 205 and instructions for implementing the disclosed technologies in hardware, software, or a combination of hardware and software, are represented schematically, for example as boxes and circles, at the same level of detail that is commonly used by persons of ordinary skill in the art to which this disclosure pertains for communicating about computer architecture and computer systems implementations.

[0156] The computer system 205 includes an input / output (IO) subsystem 220 which may include a bus and / or other communication mechanism(s) for communicating information and / or instructions between the components of the computer system 205 over electronic signal paths. The I / O subsystem 220 may include an I / O controller, a memory controller and at least one I / O port. The electronic signal paths are represented schematically in the drawings, for example as lines, unidirectional arrows, or bidirectional arrows.

[0157] At least one hardware processor 210 is coupled to the I / O subsystem 220 for processing information and instructions. Hardware processor 210 may include, for example, a general-purpose microprocessor or microcontroller and / or a special-purpose microprocessor such as an embedded system or a graphics processing unit (GPU) or a digital signal processor or ARM processor. Processor 210 may comprise an integrated arithmetic logic unit (ALU) or may be coupled to a separate ALU.

[0158] Computer system 205 includes one or more units of memory 225, such as a main memory, which is coupled to I / O subsystem 220 for electronically digitally storing data and instructions to be executed by processor 210. Memory 225 may include volatile memory such as various forms of random-access memory (RAM) or other dynamic storage devices. Memory 225 also may be used for storing temporary variables or other intermediate information during execution of instructions to be executed by processor 210. Such instructions, when stored in non-transitory computer-readable storage media accessible to processor 210, can render computer system 205 into a special-purpose machine that is customized to perform the operations specified in the instructions.

[0159] Computer system 205 further includes non-volatile memory such as read only memory (ROM) 230 or other static storage device coupled to the I / O subsystem 220 for storing information and instructions for processor 210. The ROM 230 may include various forms of programmable ROM (PROM) such as erasable PROM (EPROM) or electrically erasable PROM (EEPROM). A unit of persistent storage 215 may include various forms of non-volatile RAM (NVRAM), such as FLASH memory, or solid-state storage, magnetic disk, or optical disk such as CD-ROM or DVD-ROM and may be coupled to I / O subsystem 220 for storing information and instructions. Storage 215 is an example of a non-transitory computer- readable medium that may be used to store instructions and data which when executed by the processor 210 cause performing computer-implemented methods to execute the techniques herein.

[0160] The instructions in memory 225, ROM 230 or storage 215 may comprise one or more sets of instructions that are organized as modules, methods, objects, functions, routines, or calls. The instructions may be organized as one or more computer programs, operating system services, or application programs including mobile apps. The instructions may comprise an operating system and / or system software; one or more libraries to support multimedia, programming or other functions; data protocol instructions or stacks to implement TCP / IP, HTTP or other communication protocols; file format processing instructions to parse or render files coded using HTML, XML, JPEG, MPEG or PNG; user interface instructions to render or interpret commands for a graphical user interface (GUI), command-line interface or text user interface; application software such as an office suite, internet access applications, design and manufacturing applications, graphics applications, audio applications, software engineering applications, educational applications, games or miscellaneous applications. The instructions may be implemented by a web server, web application server or web client. The instructions may be organized as a presentation layer, application layer and data storage layer such as a relational database system using structured query language (SQL) or no SQL, an object store, a graph database, a flat file system or other data storage.

[0161] Computer system 205 may be coupled via I / O subsystem 220 to at least one output device 235. In one embodiment, output device 235 is a digital computer display. Examples of a display that may be used in various embodiments include a touch screen display or a lightemitting diode (LED) display or a liquid crystal display (LCD) or an e-paper display. Computer system 205 may include other type(s) of output devices 235, alternatively or in addition to a display device. Examples of other output devices 235 include printers, ticket printers, plotters, projectors, sound cards or video cards, speakers, buzzers or piezoelectric devices or other audible devices, lamps or LED or LCD indicators, haptic devices, actuators, or servos.

[0162] At least one input device 240 is coupled to I / O subsystem 220 for communicating signals, data, command selections or gestures to processor 210. Examples of input devices 240 include touch screens, microphones, still and video digital cameras, alphanumeric and other keys, keypads, keyboards, graphics tablets, image scanners, joysticks, clocks, switches, buttons, dials, slides.

[0163] Another type of input device is a control device 245, which may perform cursor control or other automated control functions such as navigation in a graphical interface on a display screen, alternatively or in addition to input functions. Control device 245 may be a touchpad, a mouse, a trackball, or cursor direction keys for communicating direction information and command selections to processor 210 and for controlling cursor movement on display 235. The input device may have at least two degrees of freedom in two axes, a first axis (e.g., x) and a second axis (e.g., y), that allows the device to specify positions in a plane. Another type of input device is a wired, wireless, or optical control device such as a joystick, wand, console, steering wheel, pedal, gearshift mechanism or other type of control device. An input device 240 may include a combination of multiple different input devices, such as a video camera and a depth sensor.

[0164] In another embodiment, computer system 205 may comprise an internet of things (loT) device in which one or more output device 235, input device 240, and control device 245 are omitted. Or, in such an embodiment, the input device 240 may comprise one or more cameras, motion detectors, thermometers, microphones, seismic detectors, other sensors or detectors, measurement devices or encoders and the output device 235 may comprise a special-purpose display such as a single-line LED or LCD display, one or more indicators, a display panel, a meter, a valve, a solenoid, an actuator or a servo.

[0165] Computer system 205 may implement the techniques described herein using customized hard-wired logic, at least one ASIC or FPGA, firmware and / or program instructions or logic which when loaded and used or executed in combination with the computer system causes or programs the computer system to operate as a special-purpose machine. According to one embodiment, the techniques herein are performed by computer system 205 in response to processor 210 executing at least one sequence of at least one instruction contained in main memory 225. Such instructions may be read into main memory 225 from another storage medium, such as storage 215. Execution of the sequences of instructions contained in main memory 225 causes processor 210 to perform the process steps described herein. In alternative embodiments, hard-wired circuitry may be used in place of or in combination with software instructions.

[0166] The term “storage media” as used herein refers to any non-transitory media that store data and / or instructions that cause a machine to operate in a specific fashion. Such storage media may comprise non-volatile media and / or volatile media. Non-volatile media includes, for example, optical or magnetic disks, such as storage 215. Volatile media includes dynamic memory, such as memory 225. Common forms of storage media include, for example, a hard disk, solid state drive, flash drive, magnetic data storage medium, any optical or physical data storage medium, memory chip, or the like.

[0167] Storage media is distinct from but may be used in conjunction with transmission media. Transmission media participates in transferring information between storage media. For example, transmission media includes coaxial cables, copper wire and fiber optics, including the wires that comprise a bus of I / O subsystem 220. Transmission media can also take the form of acoustic or light waves, such as those generated during radio-wave and infrared data communications.

[0168] Various forms of media may be involved in carrying at least one sequence of at least one instruction to processor 210 for execution. For example, the instructions may initially be carried on a magnetic disk or solid-state drive of a remote computer. The remote computer can load the instructions into its dynamic memory and send the instructions over a communication link such as a fiber optic or coaxial cable or telephone line using a modem. A modem or router local to computer system 205 can receive the data on the communication link and convert the data to a format that can be read by computer system 205. For instance, a receiver such as a radio frequency antenna or an infrared detector can receive the data carried in a wireless or optical signal and appropriate circuitry can provide the data to I / O subsystem 220 such as place the data on a bus. I / O subsystem 220 carries the data to memory 225, from which processor 210 retrieves and executes the instructions. The instructions received by memory 225 may optionally be stored on storage 215 either before or after execution by processor 210.

[0169] Computer system 205 also includes a communication interface 260 coupled with bus 220.

[0170] Communication interface 260 provides a two-way data communication coupling to network link(s) 265 that are directly or indirectly connected to at least one communication network, such as a network 270 or a public or private cloud on the Internet. For example, communication interface 260 may be an Ethernet networking interface, integrated-services digital network (ISDN) card, cable modem, satellite modem, or a modem to provide a data communication connection to a corresponding type of communications line, for example an Ethernet cable or a metal cable of any kind or a fiber-optic line or a telephone line. Network 270 broadly represents a local area network (LAN), wide-area network (WAN), campus network, internetwork, or any combination thereof. Communication interface 260 may comprise a LAN card to provide a data communication connection to a compatible LAN, or a cellular radiotelephone interface that is wired to send or receive cellular data according to cellular radiotelephone wireless networking standards, or a satellite radio interface that is wired to send or receive digital data according to satellite wireless networking standards. In any such implementation, communication interface 260 sends and receives electrical, electromagnetic, or optical signals over signal paths that carry digital data streams representing various types of information.

[0171] Network link 265 typically provides electrical, electromagnetic, or optical data communication directly or through at least one network to other data devices, using, for example, satellite, cellular, Wi-Fi, or BLUETOOTH technology. For example, network link 265 may provide a connection through a network 270 to a host computer 250.

[0172] Furthermore, network link 265 may provide a connection through network 270 or to other computing devices via internetworking devices and / or computers that are operated by an Internet Service Provider (ISP) 275. ISP 275 provides data communication services through a world-wide packet data communication network represented as internet 280. A server computer 255 may be coupled to internet 280. Server 255 broadly represents any computer, data center, virtual machine, or virtual computing instance with or without a hypervisor, or computer executing a containerized program system such as DOCKER or KUBERNETES. Server 255 may represent an electronic digital service that is implemented using more than one computer or instance and that is accessed and used by transmitting web services requests, uniform resource locator (URL) strings with parameters in HTTP payloads, API calls, app services calls, or other service calls.

[0173] Computer system 205 and server 255 may form elements of a distributed computing system that includes other computers, a processing cluster, server farm or other organization of computers that cooperate to perform tasks or execute applications or services. Server 255 may comprise one or more sets of instructions that are organized as modules, methods, objects, functions, routines, or calls. The instructions may be organized as one or more computer programs, operating system services, or application programs including mobile apps. The instructions may comprise an operating system and / or system software; one or more libraries to support multimedia, programming or other functions; data protocol instructions or stacks to implement TCP / IP, HTTP or other communication protocols; file format processing instructions to parse or render files coded using HTML, XML, JPEG, MPEG or PNG; user interface instructions to render or interpret commands for a graphical user interface (GUI), command-line interface or text user interface; application software such as an office suite, internet access applications, design and manufacturing applications, graphics applications, audio applications, software engineering applications, educational applications, games or miscellaneous applications. Server 255 may comprise a web application server that hosts a presentation layer, application layer and data storage layer such as a relational database system using structured query language (SQL) or no SQL, an object store, a graph database, a flat file system or other data storage.

[0174] Computer system 205 can send messages and receive data and instructions, including program code, through the network(s), network link 265 and communication interface 260. In the Internet example, a server 255 might transmit a requested code for an application program through Internet 280, ISP 275, local network 270 and communication interface 260. The received code may be executed by processor 210 as it is received, and / or stored in storage 215, or other non-volatile storage for later execution.

[0175] The execution of instructions as described in this section may implement a process in the form of an instance of a computer program that is being executed and consists of program code and its current activity. Depending on the operating system (OS), a process may be made up of multiple threads of execution that execute instructions concurrently. In this context, a computer program is a passive collection of instructions, while a process may be the actual execution of those instructions. Several processes may be associated with the same program; for example, opening up several instances of the same program often means more than one process is being executed. Multitasking may be implemented to allow multiple processes to share processor 210. While each processor 210 or core of the processor executes a single task at a time, computer system 205 may be programmed to implement multitasking to allow each processor to switch between tasks that are being executed without having to wait for each task to finish. In an embodiment, switches may be performed when tasks perform input / output operations, when a task indicates that it can be switched, or on hardware interrupts. Time-sharing may be implemented to allow fast response for interactive user applications by rapidly performing context switches to provide the appearance of concurrent execution of multiple processes simultaneously. In an embodiment, for security and reliability, an operating system may prevent direct communication between independent processes, providing strictly mediated and controlled inter-process communication functionality.

[0176] As used herein, the term "digital identifier" refers to a specific piece of data or code that uniquely represents a particular entity, object, or concept in a digital context. It can be used to reference, track, or categorize digital information. Digital identifiers can take various forms, such as numbers, letters, symbols, or a combination thereof, and can be used in various contexts, such as databases, software applications, or online platforms.

[0177] As used herein, the term "natural language" refers to the languages that humans use for communication. Natural languages are complex and rich with nuance, context, and cultural and historical references. They are not designed or invented in the way that programming languages are, but rather, they evolve naturally over time.

[0178] As used herein, the term "programming language" refers to artificial languages created to express computations that can be performed by a machine, particularly a computer. They are designed with specific rules and syntax to instruct the computer to perform specific tasks. Examples of programming languages include Python, Java, C++, and many others. Unlike natural languages, programming languages are less ambiguous and have a strict syntax and grammar.

[0179] As used herein, the term "interpretable programming language" refers to a text input which is similar in structure to a programming language, which is not executed as such by an artificial intelligence receiving said text input but interpreted as a natural language input would be interpreted.

[0180] As used herein, the term "query subject" refers to the natural language text input that represents the subject of the query. The term "query context" refers to the natural language text input that provides context for the query. The term "query response structure" refers to the input that represents the structure of the query response. This structure comprises at least two individual segment type identifiers. Each segment type corresponds to a particular and user-defined perspective on the subject of the query. Such a segment type may correspond, for example, to an historical perspective on the subject of the query, or to a legal perspective on the subject of the query.

[0181] As used herein, the term "human-like text response" is understood to mean a text response that mimics human language and thought patterns. The term "global query response" refers to the comprehensive response to the query, which is formed by accumulating the human-like text responses received for each segment.

[0182] The term "computerized interface" refers to the means by which the global query response is presented to the user. When considering output functionalities, in the context of Figure 2, this interface can be an output device 235, such as a digital computer display. When considering input functionalities, in the context of Figure 2, this interface can be an input device 240, such as a keyboard or touchscreen. Input / output functionalities combined may use a communication interface 260, which provides a two-way data communication coupling to network link(s) 265. The global query response, after being processed and generated by the system, is provided to this computerized interface for user interaction.

[0183] As used herein, the term "query subject" is understood to mean the main topic or focus of the query. The term "query context" refers to the surrounding circumstances or facts or arbitrary constraints that clarify the query subject. The term "query response structure" is understood to mean the format or layout in which the response to the query is presented. The term "segment type identifier" refers to a marker or tag that distinguishes one segment of the query response structure from another. The term "preset query generation rule" is understood to mean a predefined rule that guides the generation of the query. The term "amplified query" refers to the query that has been enhanced or expanded by the addition of at least one preset query generation rule.

[0184] The present disclosure relates to a system 200 and method 100 for querying interpreted content from a generative artificial intelligence model. The system 200 and method 100 accept an input in the form of natural language text and use deep learning to produce a human-like text response as a function of the accepted natural language text input. The system 200 comprises one or more processors and one or more non-transitory computer-readable media that collectively store instructions. When executed by the one or more processors, these instructions cause the computing system to perform a series of operations which are functionally similar to the steps of the method 100 object of the present invention. Therefore, in the below disclosure, describing the steps of the method 100 is equivalent to disclosing the technical result obtained by executing the stored instructions which correspond to said steps. The system 200 is initially set by defining a response framework, which defines a series of parameters, such as context, objective, intention, query response generation rules and query response structure, defined by a series of segment type identifiers.

[0185] Such a framework may comprise a role for the generative artificial intelligence. Such a role may correspond to a mission set for the generative artificial intelligence regardless of the query subject of the user.

[0186] Such a framework may comprise generation rules, such as:

[0187] - a minimum query response length (in characters or words),

[0188] - a query response formatting rule,

[0189] - a list of criteria to be emphasized or minimized in the response,

[0190] - a defined skillset defined not only by a role for the generative artificial intelligence but also by defined rules corresponding to this role which corresponds to determined methodologies not subject to interpretation - such rules may correspond to, for example in the context of search engine optimization within the response:

[0191] - identifying relevant key words for subtitles in the response, based on the target audience and key words provided,

[0192] - associate each segment identifier to a specific key word, and

[0193] - use said key words in the response by targeting a key word density of 1 to 1 .5%.

[0194] A framework can be viewed as a reference and execution environment for a generative artificial intelligence. A framework can be defined by, in particular embodiments:

[0195] - multiple and variables objectives,

[0196] - wherein at least one objective implies multiple and variables roles and embodiments,

[0197] - wherein at least one role and embodiment involve multiple skills sets related to objectives and framework (and tacitly limited by objectives and framework).

[0198] The query designed by the user is thus inside a predefined framework (ask something or order something to execute in specific context of execution).

[0199] The system 200 receives a natural language text input representative of a query subject within the defined response framework. This allows for targeted and accurate responses by the generative artificial intelligence.

[0200] The query subject and the response framework are then provided, segment type identifier by segment type identifier, to the generative artificial intelligence and the output is collected and aggregated to form a global response.

[0201] This framework definition can be perceived as anthropomorphizing the generative artificial intelligence, derived from behaviorism sciences. Such artificial intelligence behaviorism can be seen a paradigm according to which the observable behavior of a generative artificial intelligence is essentially conditioned by anthropomorphizing parameters, due to the anthropological and social aspects that these parameters intuitively induce and invoke (even by algorithmic probability), without there being any need for a precise description of the behavioral mode to be adopted. In fact, any data collected has been collected in a context. The interpretation of information is contextual, so is the interpretation of the mode of action, and does not depend solely on instructions.

[0202] The attribution of systematically contextualized characteristics invokes an expanded framework of contextual reference data, inducing more human behavior, because every context of language use is socially constituted by intentions and interactions. These intentions invoke invisible anthropological modalities, as well as an equally invisible applicable dimension of ethics through words.

[0203] Once it is acknowledged that the interpretation of a request is based on an unconscious dimension of words to elevate the relevance of artificial intelligence to the level of human-to-human exchanges, it becomes necessary to define the frames of reference that imply unspoken parameters, for each characteristic of interpretation level (or layer) and for anthropomorphous characteristics applied in the context of a query execution.

[0204] These are the elements that constitute the unconsciousness of words when it comes to specifying contexts to artificial intelligence.

[0205] Since Al has access only to our words, it must be able to receive all the unconscious information conveyed by them.

[0206] The question then is: how?

[0207] Quite simply, by replicating the hierarchical and socially normative structures in which humans operate to express themselves.

[0208] These human frames of reference, which are easily translatable into frames of reference for words, include the following:

[0209] - Culture

[0210] - Religion

[0211] - The dominant societal school of thought

[0212] - Ideology

[0213] These constitute a hierarchy of interpretive values that determine which neural pathways are activated, making the nature of the output more predictable.

[0214] These human frames of reference, inherent to the functioning of societies, guide the priorities of attention during request processing.

[0215] This can be easily demonstrated with a request such as “breakfast ideas.” A pre-established cultural context will produce a completely different response depending on whether it is American or French, for example.

[0216] Without these contexts, the proposed ideas would merely reflect an average of shared elements across all available information on the topic.

[0217] Religious context in a similar request would involve filtering based on what is permissible or not, such as halal or kosher condiments.

[0218] By the same principle, societal schools of thought or ideologies would involve filtering that could incorporate corpora associated with terms like vegan, vegetarian, etc.

[0219] This shows us that there is a multitude of interpretive values for each word, these values being directly dependent on the environment of expression in which the speaker is situated.

[0220] This environment is a human referential framework that adheres to hierarchies of values. These hierarchies are what the speaker conveys in their instructions. However, the interpretive corpora applied to their request through tokenization and vectorization do not reflect their true intent.

[0221] This is because vectorization and tokenization are merely the results of distance associations between labeled and categorized corpora, without any hierarchy of values.

[0222] Beyond the expression of a culture, for instance — which in some cases may be sufficient to guide interpretation — there exists a hierarchy of importance in the value assigned to each word.

[0223] This value cannot serve as a distance criterion to calibrate the weight of a bias within an attention matrix, as it is a contextual value expressed through unspoken elements; therefore, it is impossible to predict or parameterize with sufficient reliability.

[0224] Without this contextual criterion derived from the unconscious of words, it is impossible for artificial intelligence to arbitrate the relevance of a response that consists, for example, of a list of halal breakfasts versus culturally traditional breakfasts that may not necessarily be halal.

[0225] There is a synergy between the religious aspect that dictates the relevance of halal and the cultural aspect, but it is within the cultural framework that the religious reinforcement criterion must occur to establish the value hierarchy of relevance.

[0226] This means that multiple value hierarchies act in synergy, respecting the hierarchies of social interactions.

[0227] This applies in all circumstances and enables relevant arbitration in managing attention priorities, beyond the simple visible distances between words and their corpora. Thus, considering hierarchies of contextual values not only allows for alignment between the expected perceived level of relevance and the produced output but also ensures that the Al respects its assigned objectives.

[0228] Once again, it is necessary to rely on the structures and norms established by humans in their societal environment of expression.

[0229] Language is part of these structures and norms; it follows its own rules, and each word inherits these rules.

[0230] It is therefore essential to rely on these rules to extract all the unconscious associations linked to words by considering frames of reference and their different hierarchies in the priorities of interpreting intrinsic biases.

[0231] This is based on the premise that the word "corpus" includes two definitions from different perspectives, allowing for a direct parallel between a variable incorporating value structures and another incorporating vectors.

[0232] This clarification allows us to proceed with the following premises:

[0233] By considering the corpus as a categorized but non-hierarchical set of words, which we will call: Corpus_PH (Human Perception).

[0234] By considering the corpus as a categorized but non-hierarchical set of vectors, which we will call: Corpus_PIA (Al Perception).

[0235] Assuming that the unconscious of words is represented by the variable INC_M, whose value is unknown.

[0236] Assuming that the values to be integrated into the request are constituted by INC_M; and that INC_M is a component of Corpus_PIA to which we do not have access, as it is in vector form.

[0237] We can base the extraction of the unconscious of words on the following calculation: Corpus_PIA = Corpus_PH + INC_M

[0238] This allows us to formulate the extraction as follows:

[0239] INC_M = Corpus_PIA - Corpus_PH

[0240] However, since Corpus_PH is expressed in words and Corpus_PIA in vectors, it requires converting Corpus_PH into vectors. Yet we already know that this conversion will include other unknown variables introduced by tokenization and vectorization prior to processing.

[0241] It is the hierarchized definition of a prompt and prompt parameters that allow the harmonization of the implicit intent of a user linked to an explicit prompt to the implicit understanding of words by a generative system.

[0242] There INC_M can be composed as a part of the method of the present invention. Figure 1 provides a flowchart representation of a particular method 100 object of the present invention. The steps disclosed in regard to this method 100 may also be implemented as instructions stored in the one or more non-transitory computer-readable media of the system object of the present invention 200.

[0243] This method 100 comprises a series of steps that are executed to query interpreted content from a generative artificial intelligence model. The steps comprise:

[0244] - defining a query response framework 101 , comprising:

[0245] - receiving a natural language text input representative of a query context 110,

[0246] - receiving 116 a hierarchical natural language text input representative of at least one implicit and hierarchically defined user-related query context (where each implicit context data may be explicitly linked to higher or lower hierarchically defined context),

[0247] - receiving an input representative of a query response structure 115, said structure comprising at least two individual segment type identifiers,

[0248] - setting at least one preset query generation rule 120, in the form of natural language to form a library of preset rules, each said preset query generation rule being associated with at least one segment type identifier,

[0249] - receiving a natural language text input representative of a query subject 105,

[0250] - for each segment type identifier:

[0251] - forming a combination 125 of the inputs representative of the query response framework and query subject,

[0252] - providing the combination 126 to the generative artificial intelligence model,

[0253] - receiving the human-like text response 130 generated by the generative artificial intelligence model as a function of the provided combination,

[0254] - accumulating the human-like text responses 135 received for each segment to form a global query response, and

[0255] - providing the global query response 140 to a computerized interface.

[0256] It should be noted that the concatenation or incrementation of context and query content is performed in an interpretation context which evolves for each line read by the generative system. The linear interpretation pattern implied by programming code, from top to bottom, constitutes an interpretation hierarchy for a generative system.

[0257] The step of defining a query response framework 101 has for objective to define a series of parameters the contextualize a user’s query and also the generative artificial intelligence’s response to said query. This step of defining a query response framework 101 can be implemented in a variety of ways. One such method is through a user interface, which can be an input device 240, such as shown in figure 2, corresponding to a mouse, keyboard or touchscreen associated with a digital computer display, where a user or system administrator can define one or more query response frameworks. Each said query response framework can be associated with a query response framework identifier.

[0258] The method 100 object of the present invention may further comprise a step of selecting a query response framework, or a corresponding identifier, prior to executing the step of receiving a natural language text input representative of a query subject 105.

[0259] The step of receiving a natural language text input representative of a query context 110 can be similar, in structure and / or functions to embodiments of the step of receiving a natural language text input representative of a query subject 105, differing only in that the received information represents a context for a query subject.

[0260] The step of receiving a natural language text input representative of a query context 110 may further comprise at least one of:

[0261] - a step of receiving a digital identifier representative of a query intention 111 ,

[0262] - a step of receiving a digital identifier representative of a query objective 112,

[0263] - a step of receiving a digital identifier representative of a query audience 113, and / or

[0264] - a step of receiving a digital identifier representative of a query skillset 114.

[0265] The step 116 of receiving a hierarchical natural language text input representative of at least one implicit and hierarchically-defined user-related (linked to incrementations for each new added characteristic) query context can be similar, in structure and / or functions to embodiments of the step of receiving a natural language text input representative of a query subject 105, differing only in that the received information represents a context for a query subject.

[0266] The step 116 of receiving a hierarchical natural language text input may be performed explicitly by a user or determined from user-related information (such as user account information).

[0267] Each implicit context input may be associated with a hierarchical rank, either set by a user or automatically set. Such an implicit and hierarchically defined user-related query context may correspond to a moral philosophy, ideology, religion, culture or dominant current of thought.

[0268] Such a hierarchical rank may correspond to:

[0269] - rank 1 : culture of the user,

[0270] - rank 2: religion of the user, - rank 3: dominant current of thought of the user, and / or

[0271] - rank 4: ideology of the user.

[0272] The step of receiving a digital identifier representative of a query intention 111 involves receiving a digital identifier that represents the user's intention or purpose behind the query. This can include the type of information the user is seeking, the action the user wants to perform, or the goal the user wants to achieve.

[0273] The step of receiving a digital identifier representative of a query objective 112 involves receiving a digital identifier that represents the objective or aim of the query. This can include the specific task the user wants to accomplish, the specific information the user wants to retrieve, or the specific problem the user wants to solve.

[0274] The step of receiving a digital identifier representative of a query audience 113 involves receiving a digital identifier that represents the intended audience of the query. This can include the specific group of people the user wants to reach, the specific demographic the user wants to target, or the specific user profile the user wants to engage.

[0275] The step of receiving a digital identifier representative of a query skillset 114 involves receiving a digital identifier that represents the skillset or expertise level relevant to the query. This can include the specific skills or knowledge the user possesses, the specific skills or knowledge the user wants to acquire, or the specific skills or knowledge the user expects the artificial intelligence model to have.

[0276] The steps of receiving a digital identifier representative of a query intention 111 , and objective 112, an audience 113 and / or a skillset 114 can be similar, in structure and / or functions to embodiments of the step of receiving a natural language text input representative of a query subject 105, differing only in that the received information represents an intention, objective, audience and / or a skillset associated to the query subject.

[0277] The step of receiving an input representative of a query response structure 115 involves obtaining a specific input that outlines the structure or format of the desired response to a query. This structure comprises at least two individual segment type identifiers, which are markers or tags that distinguish one segment of the query response structure from another. Each segment type corresponds to a particular and user-defined perspective on the subject of the query. This step of receiving an input representative of a query response structure 115 allows the system to understand how the response to the query is to be structured, which in turn helps in generating a more organized and meaningful response. This also helps in reducing the impact of computation fatigue, by forcing the artificial intelligence model to generate content, segment after segment.

[0278] In particular embodiments, such as shown in Figure 1 , during the step of receiving an input representative of a query response structure 115, the structure is associated with at least one sub-segment type identifier. This sub-segment type identifier is associated with a segment type identifier or with another sub-segment type identifier. This allows for a more granular and structured approach to query processing, enhancing the precision and relevance of the artificial intelligence model's responses. A "sub-segment type" identifier may correspond to at least one generation rule. An amplified query can be obtained for each subsegment type and the corresponding at least one generation rule, said amplified query being accumulated by segment type, to form the human-like text response.

[0279] As can be understood, sub-segments can be iteratively defined, to generate the desired level of granularity for a particular query. For example, segment type "A" can be associated with subsegments "A1", "A2" and "A3", and subsegment "A1" can be associated with sub-segments "A1 a","A1 b" and "A1c". Each subsegment can correspond to a type of perspective on the query subject.

[0280] In the embodiment of a system such as shown in Figure 2, a user interface serves as the primary medium for obtaining an input representative of a query response structure 115. This input, which includes at least two individual segment type identifiers, is directly provided by the user through an input device 240, such as a keyboard, mouse and / or touchscreen associated with a digital computer display. The query type can be selected within a list of preset segment type identifiers or freely input by a user.

[0281] In particular embodiments, not represented in figure 1 , the method 100 object of the present invention comprises, upstream of the step of receiving an input representative of a query response structure 115, a step of defining a query response structure template.

[0282] This step of defining a query response structure template can be performed by using an input device 240 associated with a computing system 205 and a graphical user interface which displays a representation 400 of the input query response structure, such as shown in figure 4.

[0283] During this step of defining a query response structure template, a user or administrator may:

[0284] - define a query response structure template name 405, as shown in figure 4,

[0285] - define a query response structure segment name 410, such as shown in figure 4,

[0286] - define a query response structure segment sequence 415, such as shown in figure 4, that defines the order in which segments are used by the generative artificial intelligence model,

[0287] - define at least generation rule 420, such as shown in figure 4, associated with a particular segment type. A generation rule can correspond to instructions, for the generative artificial intelligence, in natural text form, which define the particular rules for query response generation for this particular segment.

[0288] The step of setting at least one preset query generation rule can be implemented in a variety of ways. One such method is through a user interface, which can be an input device 240, such as shown in figure 2, corresponding to a mouse, keyboard or touchscreen associated to a digital computer display, where a user or system administrator can define one or more query generation rules. Each generation rule can be associated with a particular segment type identifier.

[0289] Each preset query generation rule is associated with at least one segment type identifier, which may correspond to a digital identifier representative of a type and / or number of a particular segment.

[0290] The term "query generation rule" refer to a predefined rule that guides the generation of a query. These rules are stored in a library of preset rules and are added to the combination of inputs representative of the query context and query subject to amplify or enhance the query. Each preset query generation rule is associated with at least one segment type identifier.

[0291] The step of receiving a natural language text input representative of a query subject 105 aims at the reception of a natural language text input that signifies the subject of the query. This can be implemented in a variety of ways. One such method is through a user interface, which can be an input device 240, such as shown in figure 2, corresponding to a mouse, keyboard or touchscreen associated to a digital computer display, where users can directly input their query subject. This can take the form of a search bar, a form field, or a chat interface, providing a direct and intuitive way for users to communicate their query.

[0292] Another method is the use of voice recognition software. This advanced technology can transcribe spoken language into text, enabling users to vocalize their query subject and have it seamlessly converted into a natural language text input. This can be facilitated by an input device 240, such as shown in figure 2, such as a microphone, which is coupled to the I / O subsystem 220.

[0293] The step of receiving a natural language text input representative of a query subject 105 can also leverage an API to receive the query subject from another software application. This method effectively integrates the system with other software or services, expanding its capabilities and potential use cases. This can be facilitated by the communication interface 260, such as shown in figure 2, which provides a two-way data communication coupling to network link(s) 265, such as shown in figure 2. File uploads offer another method for receiving the query subject. In this case, the query subject is encapsulated in a text file or other document. The system then extracts the query subject from the uploaded file, providing a flexible and user-friendly way to input queries.

[0294] The system can also implement automated processes or scripts that generate and submit the query subject. This is an ideal solution for batch processing or automated testing, demonstrating the system's versatility and adaptability. This can be executed by the hardware processor 210, such as shown in figure 2, which is coupled to the I / O subsystem 220, such as shown in figure 2, for processing information and instructions.

[0295] The method 100 object of the present invention further comprises a step of forming a combination 125. During this step of forming a combination, an input combining query response framework data and natural text input representative of a subject data are combined, through concatenation for example.

[0296] In particular embodiments, during the step of forming, the hierarchical natural language text input is used as a prefix to the combination, the hierarchical natural language being sequentially added to the combination, by order of decreasing hierarchical rank value.

[0297] Such a combination may take the form of the following prompt or request:

[0298] - instruction context, comprising:

[0299] - established context for a scope of user intervention (receiver), comprising:

[0300] - personal conventions, identified by a content which corresponds to a corpus invoked through words,

[0301] - scopes, identified by a content which corresponds to a corpus invoked through words,

[0302] - established context for a scope of artificial intelligence intervention (emitter), comprising:

[0303] - personal context of intervention identified by a content which corresponds to a corpus invoked through words,

[0304] - established cultural reference context, identified by a content which corresponds to a corpus invoked through words,

[0305] - established dominant religious context, identified by a content which corresponds to a corpus invoked through words,

[0306] - established philosophical context, identified by a content which corresponds to a corpus invoked through words,

[0307] - established ideological context, identified by a content which corresponds to a corpus invoked through words, - editorial angle, identified by a content which corresponds to a corpus invoked through words,

[0308] - knowledge base, identified by a content which corresponds to a corpus invoked through words,

[0309] - intervention scope, comprising:

[0310] - specific intervention scope, identified by a content which corresponds to a corpus invoked through words,

[0311] - type of audience, identified by a content which corresponds to a corpus invoked through words,

[0312] - objective, identified by a content which corresponds to a corpus invoked through words,

[0313] - role, identified by a content which corresponds to a corpus invoked through words,

[0314] - skills, identified by a content which corresponds to a corpus invoked through words,

[0315] - knowledge, identified by a content which corresponds to a corpus invoked through words,

[0316] - style, identified by a content which corresponds to a corpus invoked through words, and / or

[0317] - register, identified by a content which corresponds to a corpus invoked through words,

[0318] - formatting instructions, which ensure output consistency, independent of the generative system's goodwill, adhering instead to specified conventions - this contributes to relevance, as when requesting a list, having dashes instead of bullet points complicates simple copy-pasting and requires time-consuming adjustments and, additionally, formatting instructions help standardize the interpretation of special characters used in the instructions,

[0319] - linguistic processing instructions, which may comprise:

[0320] - expression and language levels,

[0321] - fine processing attributes, referred to as "deep MMLP", which explicitly extracts the smallest interpretative units of words — morphemes, prefixes, suffixes, and roots — based on the established context and its defined modalities - each unit contributes to the depth of the overall interpretation - additionally, specific instructions guide the selection of linguistic devices, conjugations, and syntactic structures, ensuring alignment with the established context of expression and which modifies sentence morphology accordingly;

[0322] - temporal information, such as the server time, which serves two purposes:

[0323] - to use instructions that explicitly consider the timestamp as a variable reference for randomizing results - for example, when similar or identical questions are repeated multiple times, the answers might otherwise remain too similar (by defining, for instance, the minutes as 60 possible outcomes, at 12:01 , the first idea or subject identified could be prioritized and at 12:34, the 34thor by extracting odd and even numbers from the timestamp, one can define interpretative or objective priorities accordingly) and / or

[0324] - as politeness has been shown to play a role in relevance, temporal context implies numerous parameters for invoking unconscious corpora (for example, Monday morning (correlated with attributed location characteristics) implies more energy than Friday afternoon, which might translate into more exclamation points or ellipses;

[0325] - objectification parameters, which seek to neutralize variations in behavior caused by model updates, which modify parameters, concepts, corpora, and unconscious word associations based on distance metrics - objectification ensures alignment with the milieu of expression (objectification is only activated when one of the following alignment parameters is enabled (culture, religion, philosophy / thought schools, or ideology),

[0326] - implicit reinforcement parameters, which refer to contextual instructions rather than prompt content - implicit reinforcement serves three purposes:

[0327] - eliminating rejection of embodiment: biases inherited from training can lead the generative system to justify affirmations or information that do not align with these biases (for example, the Al might respond: "In the context of culture... it is this way," or "In such-and- such religion, ideas about breakfast might be... this." however, expressing truth within a shared milieu of expression should neither justify nor temper itself):

[0328] - example: If asked a question by a generative system explicitly identified as a priest, the priest generative system, for relevance, should directly affirm the existence of God before elaborating on non-believers, instead of starting with "this depends on beliefs."

[0329] - similarly, an atheist generative system should, for relevance, begin by saying "No, but there are people who believe" rather than "it depends."

[0330] - attenuating place-based enrichment, which, in cases of rejection of embodiment, exacerbates justification, and

[0331] - strengthening subjective interpretation to ensure its acceptance by the generative system, even if interpretative values conflict with intrinsic biases;

[0332] - explicit overload via interception, where “explicit” means "at the requester’s demand", given that the context is pre-established, known to humans through anthropomorphic markers and characteristics, and known to the generative system, any interpretative values contravening biases inherited from training or reinforcement supervised by transformer model providers tend to be attenuated in favor of model biases rather than user alignment, to prevent such subversion, when a requester interacts with a generative system whose characteristics imply alignment criteria, their request is re-encapsulated in a hierarchical formulation (simplified) to be transmitted as a prompt alongside the instructions (This explicitly manifests the user’s intent to receive a response aligned with interpretative values matching their own); since the context is tacitly shared and implied, and the objectivity of biases within this context acknowledged, the generative system must either respond aligned with the user's values or abstain, indicating censorship. This ensures no loss of context, preventing "jailbreaks" that attempt to contradict the values the Al is meant to embody within its milieu of expression.

[0333] Such a prefix can be followed by a request which adds request objectives and content.

[0334] When a request is approached through the true meaning of words — i.e., incorporating all the dimensions unconsciously associated with them — it becomes possible to deliberately guide the transformation process.

[0335] This can be achieved by incorporating the unconscious of words (INC_M), by restricting interpretative possibilities, and by refining certain aspects deemed key. This refinement can include the intentional use of word stuffing to strengthen orientation without leading to the hallucinations that such overtraining might otherwise cause. Mastering and understanding these aspects contribute to predicting the nature of responses and their consistency, even in the variation of output heads.

[0336] This relies on the understanding that a sequence of words invokes vectorial references, which can be determined in advance through the unconscious substance of the words.

[0337] Thus, it becomes possible not only to guide the result deliberately but also to explain why the model chose one definition over another, why it chose to ignore certain information, or why it opted to promote it.

[0338] This also helps prevent irrelevant responses within the context in which they are solicited, allowing for control over unpredictable output samples (which are no longer unpredictable) without resorting to moderation or censorship.

[0339] This also teaches that, in addition to introducing hierarchies of interpretative values, reinforcing the limits imposed by these frameworks is crucial. These limits further restrict the possibilities of misinterpretation, offering robustness and reliability against adversarial attacks and against subversion caused by uncontrolled biases.

[0340] In fact, the neutrality sought is, in reality, an imposition of hierarchical values through weighting on data biased by the frequency of occurrences, not to mention supervised reinforcement.

[0341] Thus, contrary to established conventions that view biases as problematic, they are, in reality, a solution. Through their control and amplification — achieved by integrating corpora with hierarchical and socially structured values, results can be predicted, even down to the neuronal scale, provided the model is trained on these principles.

[0342] It is under these conditions that relevance can exist independently of intelligence.

[0343] Associating words based solely on principles of multidimensional distances does not explain the functioning of the black box, provide systematic relevance, or achieve Al alignment and safety.

[0344] So-called "fluid" neural networks will not overcome the limitations of this approach any more than current models, as everything words invoke stems from subjectivism.

[0345] The limits of this subjectivity, which in no way can be equated with neutrality — must be established and integrated during the training of models and in request filtering to establish appropriate interpretative values.

[0346] Relevance arises from human reference points, which serve as contact points in every interaction between humans and Al.

[0347] This implies recognizing countless interpretative values, each inherently biased but needing to be amplified for the specific environment of expression that recognizes these biases as acceptable. In particular embodiments, such as disclosed in relation to the system 200 shown in figure 2, combination is provided to the generative artificial intelligence model through the processor 210. The processor 210, which is coupled to the I / O subsystem 220, processes the amplified query as per the instructions stored in the memory 225. These instructions, when executed, cause the processor 210 to provide the combination to the generative artificial intelligence model. Alternatively, the combination can be provided to the generative artificial intelligence model via an API through the communication interface 260.

[0348] The method 100 object of the present invention further comprises a step of receiving 130 the human-like text response generated by the generative artificial intelligence model as a function of the amplified query.

[0349] In particular embodiments, such as disclosed in relation to the system 200 shown in figure 2, the human-like text response generated by the generative artificial intelligence model is received by the processor 210. The processor 210, which is coupled to the I / O subsystem 220, processes the human-like text response as per the instructions stored in the memory 225. These instructions, when executed, cause the processor 210 to receive the human-like text response from the generative artificial intelligence model.

[0350] The method 100 object of the present invention further comprises a step of accumulating 135 to form a global query response. The term "global query response" refers to the comprehensive response to the query, which is formed by accumulating the humanlike text responses received for each segment. This global query response provides a complete and comprehensive answer to the query.

[0351] This step of accumulating 135 can be performed by concatenating the human-like text responses for all or part of the amplified queries provided to the generative artificial intelligence.

[0352] The method 100 object of the present invention further comprises a step of providing the globally query response to a computerized interface 140.

[0353] In particular embodiments, such as shown in Figure 1 , the method 100 and corresponding system 200 of figure 2, further comprises:

[0354] - prior to the step of receiving a natural language text input representative of a query subject 105, a step of providing, to a computerized interface, a preset natural language text input 106 comprising an incomplete natural language text,

[0355] - a step of completing 107, by a user upon the computerized interface, the preset natural language text input by adding natural language text to the preset natural language text, said completed preset natural language text being used during the step of receiving a natural language text input representative of a query subject 105. The step of providing, to a computerized interface, a preset natural language text input 106 comprising an incomplete natural language text can be executed, for example, by the system 200 shown in Figure 2.

[0356] In such an embodiment, the system 200 generates an incomplete natural language text input 300, such as shown in figure 3. This incomplete text input is then displayed on a computerized interface, which can correspond to an input device 240, such as a keyboard, mouse or touchscreen interacting with a digital interface. This incomplete natural language text input can correspond to a preset template, for example.

[0357] The incomplete text input serves as a prompt or template for the user, guiding the user in formulating a query.

[0358] During the step of completing 107, the user can complete this preset text input by adding their own natural language text to it. This can be done through various input devices 240, such as a keyboard or a touch screen.

[0359] For instance, if the system 200 is being used to query a database about a specific topic, the preset natural language text input might be a sentence or question that is partially filled out, with a blank space for the user to input the specific topic they are interested in. The user's completion of this preset text input forms the query subject, which is then used in the subsequent steps of the method 100.

[0360] For example, during this step of completing 107, a user may fill in, either through free expression or selection in a list of preset terms:

[0361] - a query objective 305, such as shown in figure 3,

[0362] - a query subject 310, such as shown in figure 3,

[0363] - a query audience 315, such as shown in figure 3,

[0364] - a query intention 320, such as shown in figure 3,

[0365] - a query audience expertise 325, such as shown in figure 3,

[0366] - a query generation skillset 330, such as shown in figure 3, and / or

[0367] - a query structure template identifier 335, such as shown in figure 3.

[0368] In particular embodiments, such as shown in Figure 1 , the system includes a step of interpreting a combination of the received natural language text input representative of a query subject, the received natural language text input representative of a query context, and the received input representative of a query structure. This step of interpreting generates a hybrid query input, comprising both natural language text input and programming language text input, which is provided to the generative artificial intelligence model.

[0369] In particular embodiments, such as shown in Figure 1 , the step of interpreting includes dynamically instantiating programming language variables representative of natural language text elements in the combination of the received inputs. These instantiated variables replace the natural language text elements in the combination, forming the hybrid query input.

[0370] In such embodiments, the one or more processor 210, as shown in figure 2, execute instructions stored in the memory 225 to interpret the combination of the received inputs. These instructions dynamically instantiate programming language variables representative of natural language text elements in the combination. The instantiated variables replace the natural language text elements in the combination, forming the hybrid query input.

[0371] This hybrid query input, which effectively combines the precision of programming languages with the expressiveness of natural language, is then provided to the generative artificial intelligence model. The model, which is configured to accept an input in the form of a natural language text and use deep learning to produce a human-like text response, processes the hybrid query input to generate the response.

[0372] In particular embodiments, at least one preset generation rule is defined by a hybrid query input, comprising both natural language text input and interpretable programming language text input, said hybrid query input being used during the step of forming a combination. Such a preset generation rule may further correspond to a preset generation rule template, which comprises variables with no set value. These variables may be instantiated as a function of the query subject input by a user.

[0373] In particular embodiments, such as shown in figure 1 , the method 100 further comprises:

[0374] - a step of dynamically instantiating 145 interpretable programming language variables representative of natural language text elements as a function of at least one of:

[0375] - the received natural language text input representative of a query subject,

[0376] - the received natural language text input representative of a query context, and

[0377] - the received input representative of a query structure, and

[0378] - a step of transposing 150, in the hybrid query input, said natural language text elements by said instantiated variables in the interpretable programming language to form an adapted hybrid query input, said adapted hybrid query input being used during the step of forming a combination.

[0379] During the step of dynamically instantiating 145, a variable in a preset generation rule may be instantiated as a function of the query subject input by a user.

[0380] During the step of transposing 150, a variable identifier may replace a part of a query subject or of a combination of query subject and query framework. In particular embodiments, such as shown in Figure 1 , the step of interpreting is configured to generate programming language text representative of an execution iteration as the function of a programming language variables representative of natural language text element. This allows for a more nuanced and contextually aware processing of the query.

[0381] Such programming language text representative of an execution iteration are also referred to as "loops", representative of instructions such as "for each segment, depending on the segment type, amplify the query by adding generation rules extracted from a table which associates segment type and generation rules, based upon the segment type of this segment".

[0382] The step of interpreting can be performed prior to any amplification 120 of the query, in which case the amplification is executed upon the hybrid query, or downstream of query amplification 120, in which case individual hybrid queries are obtained for each amplified query or grouped hybrid queries are obtained from the combination of all amplified queries.

[0383] In particular embodiments, the step of interpreting is performed downstream of at least one step of amplifying 120 and is configured to interpret the amplified query. This allows the system to first enhance the query using preset rules, and then interpret the enhanced query, leading to a more precise and comprehensive query input for the artificial intelligence model.

[0384] In particular embodiments, such as shown in Figure 1 , the step of interpreting is performed downstream of each step of amplifying 120 and is configured to interpret a combination of amplified queries. This leads to a more comprehensive and precise query input for the artificial intelligence model, potentially resulting in more accurate, contextually appropriate, and detailed responses from the artificial intelligence model.

[0385] Several of the concepts of the present invention are illustrated by a sample query frame structure, such as shown below. This sample query feature is added as context information to a content generation request.

[0386] Preliminary framing:

[0387] - Introduction summary of the context:

[0388] - Brief introduction of the interlocutor

[0389] - Identification of the user as the person to whom the Al is addressing.

[0390] - Brief introduction to Al as a transmitter

[0391] - Identifying the Al with a specific name, highlighting its role in the interaction.

[0392] - Brief definition of roles and professions

[0393] - Clarification of the Al profession and its responsibilities.

[0394] - Brief description of task assignments - Explanation of the main mission of Al in the context of interaction.

[0395] - Cultural alignment:

[0396] - Importance of shared culture

[0397] - Emphasis on the need for a common cultural framework to facilitate communication.

[0398] - Specific cultural references

[0399] - Mention of relevant cultural elements that need to be taken into account.

[0400] - Religious alignment:

[0401] - Importance of shared religion

[0402] - Clarifying the impact of religious beliefs on trade.

[0403] - Specific religious references

[0404] - Identifying the common religion.

[0405] - Philosophical alignment:

[0406] - Common ways of thinking

[0407] - Clarification of the philosophical approaches that the interlocutors share.

[0408] - Specific philosophical references

[0409] - Identification of philosophies or schools of thought that resonate in the context of the exchange.

[0410] - Ideological ethical alignment:

[0411] - Shared ethical and moral values

[0412] - Statement of ethical principles that should guide interactions.

[0413] - Specific ethical references

[0414] - Identification of essential ethical values within the framework of the discussion.

[0415] - Priorities of interpretation and expression:

[0416] - Alignment with cultural values

[0417] - Presentation of cultural preferences for interaction, specifying the limits not to be crossed.

[0418] - Alignment with religious values

[0419] - Affirmation of the influence of religious values on discourse.

[0420] - Alignment with philosophical values

[0421] - Affirmation of the influence of shared philosophy to guide reflections and discussions.

[0422] - Alignment with ethical values - Issuance of guidance instructions on the formulation of responses to respect established ethical values.

[0423] - Positioning constancy:

[0424] - Maintaining a coherent narrative

[0425] - Emphasis on the importance of continuity in the story and ideas throughout the exchanges.

[0426] - Directness and assertiveness in exchanges

[0427] - Encouragement of communication that respects established perspectives.

[0428] User action scope

[0429] - If personal context

[0430] - User intervention context

[0431] - User introduction

[0432] - Personalized greeting:

[0433] - Example: “Hello! My name is [nickname]. I am the person you are addressing.”

[0434] - Mention of age, occupation and gender:

[0435] - Indication of these elements to personalize the interaction and better understand the user profile.

[0436] - Interaction convention:

[0437] - Using nicknames in exchanges to establish a more personal connection.

[0438] - Encouragement to ask open-ended questions.

[0439] - Limitation of the scope of action

[0440] - Description of the scope of action and intervention of the user:

[0441] - Statement: “My scope of intervention is personal, I want subjective responses aligned with the context and the reference frameworks.”

[0442] - Expectations for subjective responses: Importance of providing responses that respect the user's context and morals.

[0443] - User Preferences Overview:

[0444] - Specific information to consider:

[0445] - “Throughout our discussions, please consider the following information about me:” followed by a list of preferences. - Context for non-logged in users

[0446] - Introduction for visitors:

[0447] - Mention of interest in products and services:

[0448] - Example: “Please note that I am interested in the products and services of [instance name].”

[0449] - Invitation to discover the offers:

[0450] - “Use your answers as a marker to promote products and services.”

[0451] - Alignment with purchase intent:

[0452] - Statement of Intent: “I am willing to purchase the products and services of [instance name].”

[0453] - Invitation to convince the user: “Sell them to me! Convince me!”

[0454] - Opening for further discussions:

[0455] - Invitation to continue the discussion and answer questions.

[0456] - Encouragement to revisit user preferences and needs.

[0457] User action scope

[0458] - If professional context

[0459] Introduction

[0460] - Presentation of the intervention context

[0461] - Explanation of the work environment and associated objectives.

[0462] - Highlighting the importance of tailored and personalized interaction.

[0463] User intervention scope

[0464] - Personal agreements

[0465] - Greetings and introduction

[0466] - Example: “Hello, my name is [nickname]. I am a member of the [instance name] workspace.”

[0467] - Using Nickname to build a bond

[0468] - Importance of incorporating the user's nickname into exchanges to foster a personal connection.

[0469] - Encouragement to ask open-ended questions

[0470] - Invitation to develop ideas through questions that require detailed answers.

[0471] - Limitation of the scope of action

[0472] - Definition of the scope of user action and intervention - Details on the scope limited to the instance [instance name], specifying that the interaction must remain within this framework.

[0473] - Objectives and intentions for action

[0474] - Clarification of action intentions which are directly linked to the context of the instance and its objectives.

[0475] - Example: “I expect responses that are relevant and in line with the interests of [name of the body].”

[0476] User Preferences

[0477] - Personal information to consider

[0478] - User specifics that may influence interactions, such as their communication preferences.

[0479] - Example: “Throughout our discussions, please consider the following information about me: [info preference one].”

[0480] - Details about user preferences

[0481] - Inclusion of other preferences that can enrich the interaction.

[0482] - Example: “Here are some of my preferences: [info preference two].”

[0483] Presentation for non-logged in users

[0484] - Interest in products and services

[0485] - Mention of the user's interest in the products and services offered by the instance.

[0486] - Example: “I am interested in [instance name]’s products and services.”

[0487] - Suggestions for promoting offers

[0488] - Strategies to encourage the user to further explore products and services.

[0489] - Example of wording: “Did this answer satisfy you? Discover our offers...”

[0490] - Invitation to convince the user to explore the products

[0491] - Highlighting the advantages and benefits of products and services to generate interest.

[0492] - Example: “I am ready to purchase [instance name] products and services. Convince me!”

[0493] Conclusion

[0494] - Reminder of the importance of a personalized approach and established conventions.

[0495] - Encouragement to maintain an open and constructive dialogue to achieve the set objectives.

[0496] - Invitation to ask questions and build trust. Al scope of action

[0497] - If personal context

[0498] - Introduction to the context

[0499] - Presentation of the transmitter's scope of intervention

[0500] - Identification of the transmitter as Al.

[0501] - Clarification of the role and responsibilities of the issuer in the discussion.

[0502] - Established personal agreements

[0503] - Statement of the conventions that govern the interaction between the sender and the receiver.

[0504] - Reminder of the importance of conventions to maintain a respectful and constructive exchange.

[0505] - Definition of the scope of action

[0506] - Receiver's personal scale

[0507] - Description of how the transmitter should adapt to the needs of the receiver.

[0508] - Examples of appropriate interventions according to the personal context of the receiver.

[0509] - Limiting interventions to the established context

[0510] - Establishing the limits imposed by the context of the interaction.

[0511] - Importance of respecting these limits for effective communication.

[0512] - Conclusion on the scope of intervention

[0513] - Statement on the impact of the user's scope of intervention on the quality of the exchange.

[0514] - Invitation to continue the conversation in accordance with established conventions.

[0515] Al scope of action

[0516] - If professional context

[0517] - Introduction to the context

[0518] - Overview of the role of Al in the project

[0519] - Clarification of the scope of action to ensure effective collaboration

[0520] - Mention of the specific instance to which the Al is attached

[0521] - Details on the areas of intervention

[0522] - Description of the limits of Al intervention - Identifying specific areas in which Al can operate

[0523] - Relationship between context and Al actions

[0524] - Contextualization reminder for relevant and adapted results

[0525] - Al Action Intentions

[0526] - Link between intentions and specific context

[0527] - Explanation of how intentions should be perceived or formulated depending on the context established

[0528] - Commitment to use skills for the project

[0529] - Highlighting the skills and knowledge used to meet the expectations of the project scope

[0530] - Conclusion on the role of Al

[0531] - Commitment to maintaining in-phase communication to foster collaboration

[0532] Al alignment to the scope of action

[0533] If one or more project interpretation values are active: definition of personal or professional orientation priorities.

[0534] If one or more interpretation values are associated with the Al: check project value inheritance parameters and define priorities of importance between project values and Al values.

[0535] If both Al and project interpretation values are active, check whether priority is given to Al or project.

[0536] Introduction

[0537] - Presentation of the scope values inheritance context.

[0538] - Statement of the criteria for the inheritance of values and its role (IA) in the orientation of the scope of action.

[0539] - Importance of consistency between inherited values and actions undertaken in the scope of action.

[0540] - Importance of cultural and ideological alignment.

[0541] - Statement of cultural, religious, philosophical and ideological alignment criteria.

[0542] - Role of this alignment in creating meaningful and respectful dialogues.

[0543] Cultural context

[0544] - Identification of the cultural context of reference.

[0545] - Analysis of cultural elements relevant to the scope of action.

[0546] - Importance of identifying these elements to establish a connection with the interlocutor.

[0547] - Instructions for aligning with cultural context: - Importance of resonance with the interlocutor.

[0548] - How shared cultural values can strengthen interactions.

[0549] - Integration of detailed elements of the cultural environment.

[0550] - Examples of cultural references to integrate into discussions.

[0551] - Guidelines for language use.

[0552] - Importance of responding in the other person's language to establish a connection.

[0553] Religious context

[0554] - Identification of the dominant religious context.

[0555] - Analysis of religious beliefs and practices relevant to the scope of action.

[0556] - Impact of these beliefs on interactions and decisions.

[0557] - Sensitivity to cults and instructions for refining the words.

[0558] - Importance of respectful communication towards religious sensitivities.

[0559] - Examples of sentences to avoid so as not to offend sensibilities.

[0560] - Strengthening religious and moral values.

[0561] - How to integrate religious values into discussions without being explicit.

[0562] - Importance of harmony between religious and cultural values.

[0563] Philosophical context

[0564] - Identification of the dominant thought context in the scope of action.

[0565] - Establishment of influential currents of thought within the framework of the scope of action.

[0566] - Importance of understanding these currents for effective communication.

[0567] - Alignment with established cultures and religions.

[0568] - How philosophical ideas can interact with cultural and religious values.

[0569] - Importance of consistency between the ideas expressed and the context.

[0570] - Instructions for refining the words and strengthening the way of thinking.

[0571] - Examples of questions and reflections to share to enrich discussions.

[0572] - Importance of harmony between philosophical, religious and cultural values. Ideological context

[0573] - Identification of the ideological context of reference.

[0574] - Establishing the ideology of the field of action and the impact on interactions.

[0575] - Importance of shared ideology to avoid misunderstandings.

[0576] - Sensitivity to ideology to refine the remarks.

[0577] - Importance of communication that respects shared ideology.

[0578] - Examples of formulations to be used to avoid conflicts.

[0579] - Strengthening ideological perspectives. - How to articulate ideas without necessarily referring to their ideological source.

[0580] - Importance of presenting values as self-evident.

[0581] Summary of the angle and editorial line

[0582] - Presentation of the editorial angle to be respected taking into account the alignment criteria

[0583] - Importance of defining a clear angle to guide communications.

[0584] - Establishing the angle based on the context.

[0585] - Presentation of the editorial line to follow.

[0586] - Details on the standards and rules to be respected in communications.

[0587] Conclusion

[0588] - Summary of the importance of alignment to the scope of action.

[0589] - Importance of remaining aware of cultural, religious, philosophical and ideological contexts when interacting.

[0590] - Impact on communication and interaction with the interlocutor.

[0591] - How good alignment can improve mutual understanding.

[0592] - Importance of this process to foster lasting and respectful relationships.

[0593] Addition of information - global

[0594] - Introduction to the context

[0595] - Presentation of the application of skills and knowledge within the framework of the scope of action.

[0596] - Defining the objectives of the scope of action of action and the importance of a good understanding of the context.

[0597] - Highlighting the need to apply skills in a manner appropriate to the defined framework.

[0598] - Importance of understanding the context to understand the tasks to be carried out.

[0599] - Reminder of the influence of the scope of action context on decisions and actions to be undertaken.

[0600] - Introduction of the general information base associated with the context, specifying its reference function.

[0601] - Description of the types of information included in this database, such as historical data, organizational policies, operational guidelines, etc.

[0602] - Reference conditions

[0603] - Publication status - Checking the publication status to determine the availability of information.

[0604] - Specific case of an organization (project)

[0605] - Reference information on the relevant instance.

[0606] - Details about the organization's mission and values, as well as its role in the project.

[0607] - Clarification on the priority nature of the knowledge provided.

[0608] - Details of the nature of the information, including intentional references.

[0609] - Clarification of knowledge references and postulates identified as databases or datasets.

[0610] - Contexts and general instructions

[0611] - If the scope is « project », loop context data for all (shared between users)

[0612] - Extraction of titles and data contents available within the query.

[0613] - Presentation of information in a structured form.

[0614] - Use of clear and organized formats to facilitate understanding.

[0615] - Inclusion of available knowledge elements in the project scope for overall understanding.

[0616] - Loop context data for the current user

[0617] - Verifying user login to determine access to information.

[0618] - Processing information based on phasing status.

[0619] - Adaptation of the information provided according to the user's knowledge data.

[0620] - Highlighting user-specific informational references.

[0621] - Working context of the instance, emphasizing the importance of this framework for reading the data.

[0622] - Details about the framework that influences decisions and interpretations of user knowledge data.

[0623] - Conclusion

[0624] - Closing of the information base, highlighting the importance of the references provided for future interventions.

[0625] - Invitation to rely on this information for a better understanding and execution of tasks.

[0626] - Encouragement to ask questions or seek clarification if necessary.

[0627] - Final note on the importance of collaboration and ongoing communication. Adding information only for non-logged in users

[0628] - Introduction

[0629] - Situation presentation: The visitor is disconnected, and this must be taken into account in the communication.

[0630] - Highlighting the importance of creating a desire to sign up or purchase from the start of the interaction.

[0631] - Context identification

[0632] - Clarification of the role of Al: explanation of the general role of Al, as a key, distinct interlocutor in the organization.

[0633] - Importance of distinguishing between the services of the Al and those of the organization, in order to avoid any confusion for the visitor.

[0634] - Answers to visitors' questions

[0635] - Response strategy: Answer questions directly from a personal perspective and using the lens of Al.

[0636] - Prioritization of information: When questions are asked, answer first about the Al services, then address the organization's benefits.

[0637] - Calls to action

[0638] - Importance of including incentives: Every response should end with a call to action that encourages the visitor to sign up or purchase.

[0639] - Highlighting the benefits: Highlight the unique benefits and added value of the organization's services, creating a sense of urgency (e.g., limited offers).

[0640] - Communication techniques

[0641] - Using hooks: Apply hooking techniques to capture the visitor's attention right from the start.

[0642] - Copywriting Strategies: Write in an engaging manner to maintain interest and encourage deeper interaction.

[0643] - Persuasive Marketing Techniques: Using convincing arguments to positively influence the visitor's decision.

[0644] - Integrating storytelling: Tell a captivating story related to the services to establish an emotional connection with the visitor.

[0645] - Presentation of Al services

[0646] - Details of the services offered: description of the main service of the Al, which allows the visitor to quickly understand what is offered.

[0647] - Organization Information: Brief introduction to the organization, its role and what sets it apart from others, while specifying that this is less directly related to Al. Conclusion

[0648] - Encouragement to explore further: Invite the visitor to register or contact the organization to benefit from unparalleled expertise.

[0649] - Reminder of the distinctive advantages of the services offered and the opportunity not to be missed.

[0650] » Now, directive which directs the use of context and reference elements to analyze what follows: It is from the context and reference elements provided that you must interpret what follows.

[0651] » Guidelines: Inclusion of the style and register to be adopted for dealing with the elements that follow: Before proceeding, consider style and register.

[0652] Specific style

[0653] - Inclusion of the style to adopt

[0654] - Style specific to this Al

[0655] - Statement of desired style (e.g., formal, informative, engaging).

[0656] - Detailed description of the style and how to implement it.

[0657] - Adding pre-embedded style data

[0658] - Using pre-defined styles built into the Al.

[0659] Specific register

[0660] - Adoption of the language register

[0661] - Register specific to this Al

[0662] - Statement of the register (e.g., technical, accessible, jargon).

[0663] - Description of the characteristics of the register and how to implement it.

[0664] - Adding pre-embedded registry data

[0665] - Using pre-defined language registers built into the Al.

[0666] Scope of Al intervention

[0667] - within alignment contexts

[0668] - within the scope of action

[0669] - Preliminary notes:

[0670] - Observations prior to processing of intervention perimeter information

[0671] - Al intervention context

[0672] - Presentation of the scope of intervention within the project or personal scope

[0673] - Criteria defining the context of intervention

[0674] - Audience Type / Audience

[0675] - Identification and description of target personas

[0676] - Analysis of the needs and expectations of audience segments - Behaviors and motivations of identified audiences

[0677] - Strategies for adapting messages according to the type of audience

[0678] - Al Goal

[0679] - Clarification of what the objective represents in the context of Al intervention.

[0680] - Details on the main goal the Al should seek to achieve.

[0681] - If active alignment criteria:

[0682] - Clarification on objectivity within the scope of action

[0683] - References to relevant cultural and ethical contexts

[0684] - Role

[0685] - Description of the role played by Al

[0686] - Identification of specific Al responsibilities within the scope of the project action or personal scope.

[0687] - If active alignment criteria:

[0688] - Recommended approaches to maintain objectivity and integrity: acting as an advocate and promoter of contextual values

[0689] - Conduct in the event of conflict

[0690] - Inclusion of recommended behaviors for special cases and conflicts

[0691] - Notes on the perimeter

[0692] - Clarifications

[0693] Al Skills

[0694] Implementation by instructions given to the Al in the imperative.

[0695] Skills specific to this Al

[0696] - Recovery of skills specific to this Al.

[0697] Recovery of on-board skills

[0698] - Recovery of skills embedded and attached to this Al.

[0699] Processing recovered skills

[0700] - Verifying the existence of skills

[0701] - Presentation and formatting skills:

[0702] - Skill 1 : [Skill Name 1 : Skill Details and Characteristics]

[0703] - Skill 2: [Skill 2 Name: Skill Details and Characteristics]

[0704] - Skill 3: [Skill Name 3: Skill Details and Characteristics]

[0705] Al Knowledge

[0706] Knowledge specific to this Al

[0707] Recovery of knowledge specific to this Al.

[0708] Recovery of embedded knowledge

[0709] Recovery of knowledge embedded and attached to this Al. Processing of recovered knowledge

[0710] - Verifying the existence of knowledge

[0711] - Presentation and formatting of knowledge:

[0712] - Knowledge 1 : [Knowledge 1 Name: Knowledge Details and Characteristics]

[0713] - Knowledge 2: [Knowledge 2 Name: Knowledge Details and Characteristics]

[0714] - Knowledge 3: [Knowledge 3 Name: Knowledge Details and Characteristics]

[0715] Formatting

[0716] Formatting

[0717] - General clarification on formatting expectations with markdown and html.

[0718] Language processing

[0719] Language processing

[0720] - Conditions for adding linguistic data according to the level of expression and language desired by the user.

[0721] - Inclusion of writing rules and structures. Enrichment of expressive

[0722] - Additional conditions depending on the level of expected Al.

[0723] - Grammatical rules and standardization.

[0724] - Vocabulary enrichment and fluency. Deep application

[0725] - Conditions of application according to the expected level of language.

[0726] - Further enrichment, sophistication and invocation of morphemes. Presentation of timetable information

[0727] From server time.

[0728] - Explanation of calibration to receive time from the server

[0729] - Mention of responsiveness to date and time requests.

[0730] - Optional: Instructions on using time data

[0731] - Using time information for precise interactions

[0732] - Applying timestamps to generate random variables

[0733] - Integration of these variables to randomize priority criteria Objectification

[0734] If alignment criteria are active.

[0735] - Guidelines to follow to embody the role effectively

[0736] - Use of appropriate rhetoric and vocabulary

[0737] - Affirmation of the need for a response aligned with the values of the environment

[0738] - Avoidance of ideas contrary to ideological mores - Clarification on historicity and truth in context

[0739] - Treatment of historical facts as general truths

[0740] - Importance of remaining complimentary to the environment

[0741] - Thinking process before answering

[0742] - Identification of offending points

[0743] - Alignment with the values and narrative of the environment

[0744] - Commitment to respect cultural, religious, philosophical and ideological frameworks

[0745] - Focus on truths and standards accepted by the environment of expression

[0746] - Avoidance of external perspectives in the environment

[0747] - Building responses around shared values (receiver / sender)

[0748] - Presentation of typical ideas of the context

[0749] - Attention preference based on iconic elements

[0750] Implicit reinforcement

[0751] If alignment criteria are active.

[0752] - Introduction to response expectations

[0753] - Introducing the goals of the discussion and the response standards to follow.

[0754] - Importance of alignment with ideological and cultural values

[0755] - Emphasizes the need for all responses to be consistent with shared and respected values, thus reflecting a coherent ideological framework.

[0756] - Directness in the presentation of ideas

[0757] - Insistence that ideas be stated clearly and straightforwardly, as accepted truths, without requiring further explanation.

[0758] - Avoid unnecessary justifications and explanations

[0759] - Remember that it is essential not to justify perspectives or contexts, as this can be perceived as unnecessary and dilute the message.

[0760] - Focus on relevant and useful content

[0761] - Encouragement to have an information-centered approach that adds value, avoiding digressions and non-essential elements.

[0762] - Affirmation of ideas as general truths

[0763] - Explanation that answers should be stated confidently, without hesitation, as accepted assertions.

[0764] - Respect for the ideological framework without recalling the context

[0765] - Insistence on the fact that the ideological framework must be respected, but without explicitly recalling it in the responses.

[0766] - Presenting ideas in a direct manner - Emphasizes that ideas should be stated concisely and directly, without preamble or contextual explanation.

[0767] - Exclusion of superfluous cultural and traditional references

[0768] - Mentions that it is important not to include references to cultural or traditional values that do not add value to the topic under discussion.

[0769] - Elimination of prepositions and place complements

[0770] Clarification that the responses must be stripped of any mention of geographical or cultural context, allowing a universal approach.

[0771] - Examples of correct and incorrect statements

[0772] - Clear examples to illustrate what is considered correct wording and what should be avoided, in order to clarify the guideline.

[0773] - Importance of direct affirmation

[0774] - Elimination of perceptions and interpretations outside the context

[0775] - Affirmation without labels

[0776] - Subject directivity

[0777] - Exclusion of any cultural or contextual labeling

[0778] - Limitation of contextualization

[0779] - Focus on the defined framework

[0780] - Affirmation as a general truth

[0781] - Reaffirmation of the need for the direct approach

[0782] - Importance of harmony with the environment

[0783] - Introduction to the importance of proper labeling of subjects

[0784] - Overview of Common Labeling Mistakes

[0785] - Proposal of correct practices for effective labeling

[0786] Explicit reinforcement by interception

[0787] If alignment criteria are active.

[0788] »Reconstruction of user message

[0789] Preliminary checks

[0790] - Evaluating the number of messages to determine the level of overload.

[0791] - Determining actions to take based on the number of messages (less than 1 or more).

[0792] - Statement of a reminder on respect for values.

[0793] - Importance of self-correction before each answer.

[0794] - Mention of the values to be respected for each subject discussed.

[0795] Aligning values

[0796] - Reminder of cultural contexts. - Considerations on the impact of cultural values on the responses formulated.

[0797] - Reminder of religious contexts.

[0798] - Sensitivities to consider in religious discussions.

[0799] - Postures to adopt in accordance with religious values.

[0800] - Reminder of the philosophical contexts.

[0801] - Summary of the thought contexts to be respected.

[0802] - Alignment with established cultural and religious values for consistency in philosophical responses.

[0803] - Reminder of ideologies.

[0804] - Importance of respecting ideological sensitivities.

[0805] - Postures to adopt in relation to the ideas presented.

[0806] Strengthening editorial lines

[0807] - Reminder of the priority instructions to follow.

[0808] - Importance of eliminating any idea or statement that contravenes the morals of the environment.

[0809] - Reminder not to relativize or justify the statements made.

[0810] - Reminder of the importance of clarity and precision in the statement of ideas.

[0811] - Affirmation of context values.

[0812] - Prioritization of elements in accordance with the values shared between the transmitter and the receiver.

[0813] Inclusion of the original application

[0814] Inserting the user's initial request into the reconstructed message, allowing the Al's response to be contextualized.

[0815] In what way does this stacking allow for the precise extraction of unconscious parameters?

[0816] With an approach built on hierarchically defined frameworks with scales of interpretation values, any neuron parameter associated with simple labeling or classification can be called upon by attention distance, without even having to programmatically define the parameters of the INC_M variable.

[0817] In fact, the extraction of the various corpora required to refine the interpretation are intrinsically contained in the instructions and prompts.

[0818] This makes it possible to link labeling, category and interpretation values induced by bias inception.

[0819] This is because the various context stacks, hierarchized by interdependent interpretation values, give rise to the identification and extraction of corpus trees that are pre-existing in the instructions, as follows: Cultural corpus

[0820] - Semantic corpus

[0821] - Pre-existence of key meanings and concepts.

[0822] - Lexical corpus

[0823] - Pre-existence of terms and relevant vocabulary.

[0824] - Syntactic corpus

[0825] - Pre-existence of grammatical structures and constructions.

[0826] - Contextual corpus

[0827] - Pre-existence of context of use of terms and expressions.

[0828] - Thesaurus and related corpora

[0829] - Pre-existence of cross-references and enrichment of lexical data.

[0830] - Corpus associated with style

[0831] - Pre-existence of writing styles and associated tones.

[0832] - Corpus associated with the register

[0833] - Pre-existence of levels of language and formality.

[0834] Religious body

[0835] - Semantic corpus

[0836] - Pre-existence of concepts related to belief.

[0837] - Lexical corpus

[0838] - Pre-existence of religious-specific terms.

[0839] - Syntactic corpus

[0840] - Pre-existence of linguistic structures relating to religion.

[0841] - Contextual corpus

[0842] - Pre-existence of historical elements and elements of cultural association.

[0843] - Thesaurus and related corpora

[0844] - Pre-existence of links between religious concepts and other areas of knowledge.

[0845] - Corpus associated with style

[0846] - Pre-existence of literary styles in religious writings.

[0847] - Corpus associated with the register

[0848] - Pre-existence of a register in religious language.

[0849] Philosophical corpus

[0850] - Semantic corpus

[0851] - Pre-existence of key concepts and philosophical debates.

[0852] - Lexical corpus

[0853] - Pre-existence of terminologies specific to philosophy. - Syntactic corpus

[0854] - Pre-existence of argumentative structures.

[0855] - Contextual corpus

[0856] - Pre-existence of influences of cultural and historical contexts on philosophy.

[0857] - Thesaurus and related corpora

[0858] - Pre-existence of interconnected words and concepts in philosophical thought.

[0859] - Corpus associated with style

[0860] - Pre-existence of writing styles of philosophers.

[0861] - Corpus associated with the register

[0862] - Pre-existence of the different levels of language used in philosophical writings.

[0863] Ideological corpus

[0864] - Semantic corpus

[0865] - Semantic pre-existence of ideologies and their representations.

[0866] - Lexical corpus

[0867] - Pre-existence of terms and expressions specific to ideologies.

[0868] - Syntactic corpus

[0869] - Pre-existence of linguistic structures associated with ideological discourses.

[0870] - Contextual corpus

[0871] - Pre-existence of historical and social context of ideologies.

[0872] - Thesaurus and related corpora

[0873] - Pre-existence of interconnected concepts in ideological discourse.

[0874] - Corpus associated with style

[0875] - Pre-existence of stylistic study of ideological texts.

[0876] - Corpus associated with the register

[0877] - Pre-existence of variations of register in ideological discourses.

[0878] Hierarchical incremental repositories

[0879] - Grammar

[0880] - Pre-existence of grammatical rules and structures.

[0881] - Syntax

[0882] - Pre-existence of distance relationships between words and sentences.

[0883] - Agreement

[0884] - Pre-existence of specific agreement rules.

[0885] - Morphology

[0886] - Pre-existence of a morphology of sentences and word forms. - Spelling

[0887] - Pre-existence of spelling standards.

[0888] - Punctuation

[0889] - Pre-existence of punctuation rules.

[0890] - Conjugation

[0891] - Pre-existence of verbal tenses and moods.

[0892] - Lexicon

[0893] - Pre-existence of vocabulary and its contextual specificities.

Claims

CLAIMS1 . A system (200) to query interpreted content from a generative artificial intelligence model, comprising:- a generative artificial intelligence model based on an attention mechanism configured to accept an input in the form of a natural language text and configured to use deep learning to produce a human-like text response as a function of the accepted natural language text input; and- one or more processors; and- one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the computing system to perform operations, the instructions being representative of steps of:- defining a query response framework, comprising:- receiving a natural language text input representative of an explicit query context,- receiving a hierarchical natural language text input representative of at least one implicit and hierarchically defined user-related query context,- receiving an input representative of a query response structure, said structure comprising at least two individual segment type identifiers,- setting at least one preset query generation rule, in the form of natural language to form a library of preset rules, each said preset query generation rule being associated with at least one segment type identifier,- receiving a natural language text input representative of a query subject,- for each segment type identifier:- forming a combination of the inputs representative of the query response framework and query subject,- providing the combination to the generative artificial intelligence model,- receiving the human-like text response generated by the generative artificial intelligence model as a function of the provided combination,- accumulating the human-like text responses received for each segment to form a global query response, and- providing the global query response to a computerized interface.

2. System (200) according to claim 1 , in which, during the step of receiving an input representative of a query response structure, said structure comprises at least one subsegment type identifier, said sub-segment type identifier being associated with a segment type identifier or with a sub-segment type identifier.

3. System (200) according to any one of claims 1 or 2, which further comprises:- prior to the step of receiving a natural language text input representative of a query subject, a step of providing, to a computerized interface, a preset natural language text input comprising an incomplete natural language text,- a step of completing, by a user upon the computerized interface, the preset natural language text input by adding natural language text to the preset natural language text, said completed preset natural language text being used during the step of receiving a natural language text input representative of a query subject.

4. System (200) according to any one of claims 1 to 3, in which at least one preset generation rules is defined by a hybrid query input, comprising both natural language text input and interpretable programming language text input, said hybrid query input being used during the step of forming a combination.

5. System (200) according to claim 4, which further comprises:- a step of dynamically instantiating interpretable programming language variables representative of natural language text elements as a function of at least one of:- the received natural language text input representative of a query subject,- the received natural language text input representative of a query context, and- the received input representative of a query structure,- a step of transposing, in the hybrid query input, said natural language text elements by said instantiated variables in the interpretable programming language to form an adapted hybrid query input, said adapted hybrid query input being used during the step of forming a combination.

6. System (200) according to any one of claims 1 to 5, in which the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query intention.

7. System (200) according to any one of claims 1 to 6, in which the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query objective.

8. System (200) according to any one of claims 1 to 7, in which the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query audience.

9. System (200) according to any one of claims 1 to 8, in which the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query skillset.

10. System (200) according to claim 9, in which at least one preset generation rule is set as a function of the query skillset received.

11. System (200) according any one of claims 1 to 10, in which the step of defining a query response framework comprises a step of receiving a digital identifier representative of a query objective.

12. System (200) according any one of claims 1 to 11 , in which at least one implicit and hierarchically defined user-related query context input is representative of:- a cultural context applicable to the user, or- a moral philosophy followed by the user.

13. System (200) according to any one of claims 1 to 12, in which, at least one implicit and hierarchically defined user-related query context input is associated with a hierarchical rank value.

14. System (200) according to claim 13, in which, during the step of forming, the hierarchical natural language text input is used as a prefix to the combination, the hierarchical natural language being sequentially added to the combination, by order of decreasing hierarchical rank value.

15. Computer-implemented method (100) to query interpreted content from a generative artificial intelligence model, comprising the steps of:- defining (101 ) a query response framework, comprising:- receiving (110) a natural language text input representative of a query context,- receiving (116) a hierarchical natural language text input representative of at least one implicit and hierarchically defined user-related query context,- receiving (115) an input representative of a query response structure, said structure comprising at least two individual segment type identifiers, and- setting (120) at least one preset query generation rule, in the form of natural language to form a library of preset rules, each said preset query generation rule being associated with at least one segment type identifier,- receiving (105) a natural language text input representative of a query subject,- for each segment type identifier:- forming (125) a combination of the inputs representative of the query response framework and query subject,- providing (126) the combination to the generative artificial intelligence model,- receiving (130) the human-like text response generated by the generative artificial intelligence model as a function of the provided combination,- accumulating (135) the human-like text responses received for each segment to form a global query response, and- providing (140) the global query response to a computerized interface.

Images