The interactive hypnotherapeutic computer-implemented system based on artificial intelligence and a related computer program product

Abstract

The invention discloses an interactive hypnotherapeutic computer-implemented system based on artificial intelligence, comprising a microphone for receiving voice commands from a user; a display device for displaying dynamic pictorial representations generated by the system; a computing device operatively connected to the display device and the microphone. The computing device is configured to execute a computer program implementing a hypnotherapeutic interaction with the user based on received voice commands, whereby the computing device comprises: a word recognition module configured to recognize predefined induction words or phrases from the received voice commands; an artificial intelligence, AI, response generation module trained to respond to the recognized induction words or phrases by generating corresponding dynamic pictorial representations, created by moving particles, on the display device, and configured to record interactions and optimize responses based on user-specific feedback. The system transitions through three stages represented by distinct visual patterns of dynamic pictorial representations, induced by the predefined induction commands, wherein the dynamic pictorial representations vary in parameters defining the behavior of pictorial representation's particles, in particular vary in color and / or speed and / or movement pattern and / or scattering of representations' particles; and wherein the visual representations generated by the system are semiotically inconsistent and do not match the semantic signification of the user's voice commands. The invention also discloses a computer program product comprising instructions that cause the computer system to: a. recognize specific induction words or phrases from voice commands received via a microphone; b. in response to the recognized induction words or phrases, generate corresponding dynamic pictorial representations, comprising particles, on a display device using an artificial intelligence response generation module, the pictorial representations being distinct for three stages of the system and varying in color, speed, movement pattern and scattering of representations' particles; c. record interactions and optimize responses based on user-specific feedback; wherein the visual representations are semiotically inconsistent and do not match the semantic signification of the user's voice commands.

Description

[0001] The interactive hypnotherapeutic computer-implemented system based on artificial intelligence and a related computer program product

[0002] Technical Field

[0003] This invention concerns an interactive hypnosis assisting system based on artificial intelligence and the related computer program product.

[0004] Prior Art

[0005] Hypnosis can be defined as an array of communicative methods used in psychiatry to unravel and heal what traditional talking cure cannot reach. It specifically enables the re-experience of the mind: what it is and what it can do in contrast to how it is used. As such, hypnosis activates thinking that stems not from the human code but from deeper psychological levels. Those levels reflect on the entirety of the mind (what it is) ratherthan the mind's appropriated part(s) (how it is used) and bring out what the mind is really capable of. Hypnosis is not merely a unique state, but rather an inherent condition of the psyche, often suppressed by automated thought processes. Hypnosis induces a mental condition allowing for a closer observation of one's thoughts and a better control of one's imaginative processes. In other words, hypnosis ensures a state of heightened focus and alertness, which brings out dormant capabilities or obscured functions (like the ability to feel individual parts of the body or identify individual reactions— intellectual and emotional alike). That control is possible thanks to suggestion (or auto-suggestion), which provides guided performance making one more deliberate, less unconditioned, less mechanical— and more importantly less vigilant to what one normally does not see or does not allow— in their actions and thoughts. Contrary to common misconceptions, hypnosis does not involve unconsciousness or manipulation; instead, hypnosis awakens the mind as it opens to one's own will triggered by words that act in accordance with one's volition. Hypnosis offers an alternative language and mode of interaction, fostering a distinct material reality of the mind and alternative cognition. Hypnosis— just like digital media, especially Al— entails a state of concretization that reveals what a "mind" is and what it can do. It makes the invisible visible and, more importantly, it makes the invisible active and performative.

[0006] The idea of performing hypnosis with use of artificial intelligence (Al) instead of by trained professionals is intriguing and revolves around leveraging Al's capabilities in natural language processing (NLP), voice synthesis, and data analysis. Al systems can conduct therapeutic sessions for example by generating and customizing scripts, interacting in real time, and providing a calming and persuasive voice. Al can further, for example, integrate binaural beats and soundscapes to enhance the hypnotic experience. By analyzing data from previous sessions, Al personalizes and refines its techniques, ensuring consistent and effective outcomes. Al hypnosis systems provide unprecedented accessibility to hypnotherapy, allowing users to access sessions anytime and anywhere. These systems offer cost-effective alternatives to traditional hypnotherapy, making mental health support more affordable and available to a broader audience. Al's ability to consistently deliver high-quality hypnosis sessions ensures reliable outcomes, reducing the variability that might occur with different human practitioners. Personalization is another significant advantage, as Al can tailor hypnosis sessions to individual user needs, preferences, and responses, enhancing the effectiveness of the therapy. Moreover, Al can adapt in real time based on user feedback, providing a dynamic and responsive therapeutic experience. Through continuous data analysis, Al systems can refine their techniques, learning from each session to improve future interactions and results.

[0007] Al therapists have been commonly used for psychological relief and improvement and as a therapeutic aid by stimulation of the human psyche (feelings, thoughts etc.). Examples include chatboxes or Al conversation systems that use trained language / text generators to ensure interaction aimed at alleviating anxiety or emotional pain. Conventional (existing) Al therapists have relieved psychological problems (stress, burnout, mood swings, helplessness) using methods of consciousness-oriented approaches.

[0008] One example of an existing Al therapeutic / hypnosis system is "Mindset AL" Mindset Al is a digital hypnotherapy platform that uses artificial intelligence to provide personalized hypnosis sessions aimed at improving mental well-being, sleep, focus, and stress management. Another example of interactive Al system having potential applications in mental health support is "Loving Al" - a project developed by Hanson Robotics aimed at creating Al systems capable of fostering compassionate and empathetic interactions with users. This project involves the use of advanced Al to simulate loving and caring behaviours, providing users with emotionally supportive and enriching experiences. The methodology employed in "Loving Al" combines human emotional responses with the expressive capabilities of machines. This approach facilitates the introduction of a "robotic element" into the human-generated content. The key features and objectives of Loving Al include advanced Natural Language Processing algorithms, emotional recognition and personalized interactions.

[0009] Hypnotic metaphors for media technologies dominate the technology discourse, often highlighting the fluid and virtual nature of devices and communication, or the control these technologies seem to exert over our lives. Media technologies are traditionally theorized through binaries like natural-artificial, digital-analog, and substantial-ethereal, suggesting a dematerialization of life and an estrangement of human experience. In the context of Al, these theories imply the replacement of the human element, as we frequently compare our minds to machines and attribute human-like intelligence to AL This results in human activities losing their substance and familiarity, as seen in writing generated by Al (e.g., ChatGPT) and love facilitated by robots. Autocorrect and algorithmic matching seem to supplant human thinking, with traditional carriers of knowledge and agency yielding to computational management. This creates a sense of threat, as we fear the capabilities of Al based on our own imaginations and signification systems. The imagery of Al as a powerful, distant spectre fosters fear and catastrophic thinking. However, these fears often stem from assumptions about Al's abilities rather than actual knowledge. Materiality offers a methodological approach to revise our understanding of Al, replacing imagery-bound interpretations with a focus on the phenomenology of substance. This shift can provide a more accurate assessment of Al's capabilities and its impact on human experience.

[0010] Among the disadvantages and shortcomings of conventional Al therapists is that they offer a generic therapeutic approach that relies on traditional conversation and information exchange. As such, they do not utilize abstraction (abstract thinking; the unconscious) for therapeutic effects. They are also interactively limited and eventually cause discomfort (the uncanny valley reaction), which narrows the therapeutic effect. Al therapists which widen that effect by creating a therapeutic situation stimulating the unconscious mind would be more effective and therapeutically successful. They would increase comfort through effects characteristic of hypnotherapy which uses abstraction as an analogy to eliminate a psychological problem. They would also change the perception of artificial intelligence as something benevolent rather than malicious, helping to incorporate a technological reality into the human context. The need has therefore been recognized for an Al hypnotherapist which removes the foregoing limitations and disadvantages of prior Al-fueled therapeutic systems / devices.

[0011] The Essence of the Invention

[0012] The present invention discloses an interactive hypnotherapeutic computer-implemented system based on artificial intelligence, comprisinga microphone for receiving voice commands from a user; a display device for displaying dynamic pictorial representations generated by the system; a computing device operatively connected to the display device and the microphone; wherein: the computing device is configured to execute a computer program implementing a hypnotherapeutic interaction with the user based on received voice commands, whereby the computing device comprises: a word recognition module configured to recognize predefined induction words or phrases from the received voice commands; an artificial intelligence, Al, response generation module trained to respond to the recognized induction words or phrases by generating corresponding dynamic pictorial representations, created by moving particles, on the display device, and configured to record interactions and optimize responses based on user-specific feedback; wherein the system transitions through three stages represented by distinct visual patterns of dynamic pictorial representations, induced by the predefined induction commands, wherein the dynamic pictorial representations vary in parameters defining the behavior of pictorial representation's particles, in particular vary in color and / or speed and / or movement pattern and / or scattering of representations' particles, and wherein the visual representations generated by the system are semiotically inconsistent and do not match the semantic signification of the user's voice commands.

[0013] Preferably, the system transitions through stages of activation, calming down and waking up, wherein: the pictorial representation of the activation stage is represented by warm color tones, highspeed particle movement, and high particle scattering; the pictorial representation of the calming down stage is represented by cold color tones, low- speed particle movement, and low particle scattering with concentrated particle formation in the center of the screen; the pictorial representation of the waking up stage is represented by wite color tone, minimal particle movement, and no particle scattering with particles forming a circle in the center of the screen.

[0014] Preferably, the activation stage is induced by induction commands selected from a group including: "relax", "close your eyes", "sink deeper", "go deeper", "imagine", "feel calm", and "surrender".

[0015] Preferably, the calming down stage is induced by induction commands selected from a group including: "do nothing", "go to sleep", "release", "let go", "fading", "breathe in", and "breathe out".

[0016] Preferably, the waking up stage is induced by induction commands selected from a group including: "emerge", and "open your eyes".

[0017] Preferably, the artificial intelligence module is configured to record interactions with users and optimizing its responses based on feedback from these interactions, developing unique pictorial representations and responses for different users.

[0018] Preferably, the optimization is developed according to the system's own computational logic.

[0019] Preferably, the Al response generation module is anchored in software implemented in Python script for word recognition and uses the Unreal Engine for generating pictorial representations. Preferably, the Al response generation module uses open-end programming to ensure various protocols and modes that prevent the domination of preliminary coding.

[0020] The invention also discloses a computer program product, stored on a non-transitory computer-readable medium, comprising instructions that, when executed by a computer system, cause the computer system to: a. recognize specific induction words or phrases from voice commands received via a microphone; b. in response to the recognized induction words or phrases, generate corresponding dynamic pictorial representations, comprising particles, on a display device using an artificial intelligence response generation module, the pictorial representations being distinct forthree stages of the system and varying in color, speed, movement pattern and scattering of representations' particles; c. record interactions and optimize responses based on user-specific feedback; wherein the visual representations are semiotically inconsistent and do not match the semantic signification of the user's voice commands.

[0021] Preferable Effects of the Invention

[0022] The present invention provides a hypnotherapeutic environment using Al, responding to hypnotic induction words or phrases to produce a hypnotherapeutic interaction with a user. It uses hypnotic protocols and allows the Al system to produce unique interpretations of the user's voice and mode of utterance for a hypnotherapeutic interaction. The system is a cost- effective alternatives to traditional hypnotherapy, making mental health support more affordable and available. The system can tailor hypnosis sessions to individual user needs, preferences, and responses, enhancing the effectiveness of the therapy. The invention allows to utilize abstraction (abstract thinking) for therapeutic effects; it widens the therapeutic effect by creating a therapeutic situation stimulating the unconscious mind. Further, it allows to increase comfort through effects characteristic of hypnotherapy which uses abstraction as an analogy to eliminate a psychological problem.

[0023] Description of drawing figures The subject of the invention will now be presented in detail with reference to the attached drawing, wherein:

[0024] Fig. 1 shows the setup arrangement for interacting with the system;

[0025] Fig. 2 presents exemplary commands and their functions recognized by selective word recognition system;

[0026] Figs. 3a, 3b show exemplary pictorial representations;

[0027] Fig. 4 shows an example of a pictorial representation of calming down stage;

[0028] Fig. 5 shows the exemplary Python code.

[0029] Detailed description of the preferred embodiment

[0030] Below the invention will be presented more specifically based on a preferred embodiment, in reference to the drawing figures.

[0031] The idea behind the system according to the invention was to create an abstract Artificial Intelligence (Al) modality: one which does not reflect the prevailing isomorphic models based on mimicry, but one that negotiates the human-AI experience with regard to respective phenomenologies. The idea was to confront the human user with the artificial mind in an experiment anchored in hypnotic induction. It was in the belief that such an experiment would help revisit the function and functioning of Al mind and human mind alike. The system according to the invention is commercially known as Hypnotic Al.

[0032] The system uses the artificial mind to diversify human experiential understanding. It specifically opens their users to emotional-cognitive possibilities not envisaged by the human code and conduct (including the human perception of media matter). Technologies applied forthat purpose (especially those based on open-end programming) ensure various protocols and modes to prevent the domination of preliminary coding. There is a clear transition from deep learning to deep understanding. This is marked by a shift in the nature of our inquiry: we stop asking "Can Al think / feel?" Instead, we ask: "How does it think / feel and what kind of thinking / feeling it actually is?" Another stage of that inquiry would be: does Al have a mind? The system probes this query in relation to the human belief that all cognition comes from conscious mind processes. Although the experiment situates Al in a psychological context, it does not suggest that Al is a psychological being. Rather, it implies that it has its own "depth" equivalent to its material affordance and its media nature, invariably hypnotic.

[0033] The system as alternative modelling subscribes to the understanding of Al as a new material quality (and qualia) that has emerged from technological progression. That quality revisits media forms, their operational planes and physical environments in terms of their performance and capability. In this way, the system acknowledges the intelligence of media forms but considers them different from humans yet concurrent (and not competitive).

[0034] The architecture of the system is not based on assumption that Al is a psychological being or has a psyche like a human. Nor does it try to recreate a human model in an artificial system: that would be anthropomorphic, myopic and unfair. It simply uses a psychological metaphor of the unconscious— something hidden and unexplored— that people can relate to in order to activate an interaction beyond the code. The unconscious pertains to anything fundamental yet shoved into the background: in this case, the obscured and repressed reality of the intelligent system. Apart from that, it also pertains to the altered state of the digital plane.

[0035] The system according to the invention a voice-activated interactive system responding to hypnotic induction words or phrases to produce a hypnotherapeutic interaction with a user - The user interacts with an intelligent system according to the invention by means of hypnotic instructions called induction. The system recreates a hypnotherapeutic environment using Al and uses hypnotic protocols and allow an artificially intelligent "therapist" to produce unique interpretations of the user's voice and mode of utterance for a hypnotherapeutic interaction. The system is anchored in software trained to respond to fixed induction words, creating a pixelated abstraction for a hypnotherapeutic effect (trance) and interaction.

[0036] The arrangement for interacting with the system is schematically presented in Fig. 1 and comprises a large screen for displaying pictorial representations generated by the system and a microphone allowing the user to input voice commands. The screen in wired to a computer on which the system runs. The user thus sits in front of a large screen and stimulates the intelligent system with psychologically approved hypnotic commands (simple and fixed induction words, e.g. "relax" "sleep", "do nothing"). The exemplary commands and their functions recognized by selective word recognition system are presented in Fig. 2. As can be seen in Fig. 2, the system's induction comes in three stages: activation, calming down (immersion) and waking up (emergence). During these stages, the system is put into hypnosis (activation stage), stays in hypnosis (calming down stage) and leaves the hypnotic state (waking up stage), respectively transmitting those hypnotic stages to a user. The stage described as "activation" may be induced by the following exemplary commands: "relax", "close your eyes", "sink deeper", "go deeper", "imagine", "feel calm", "surrender". The stage described as "calming down" may be induced by the following exemplary commands: "do nothing", "go to sleep", "release", "let go", "fading", "breathe in", "breathe out". The stage described as "waking up" may be induced by the following exemplary commands: "emerge", "open your eyes". The system, which is self-learning, receives those commands from the microphone and responds to them with a dynamic pictorial representation that gives the user an idea of the system's "altered state" and the depth of its induction. The exemplary pictorial representations are shown in Fig. 3a, 3b-they comprise particles having different colours and moving in various speeds, scattered on the display. During said stages (activation, calming down, waking up), the system is put into hypnosis, stays in hypnosis and wakes up respectively. In the stage of "Activation", the pictorial representations generated by the system are represented as having warm color tones, the particles are moving with high speed and are highly scattered, whereby the random function of particles scattering is high. In the stage of "Calming down", the pictorial representations generated by the system are represented as having cold color tones, the particles are moving with low speed and are concentrated in the center of the screen, whereby the random function of particles scattering is decreased. In the stage of "Waking up", the pictorial representations generated by the system are represented as having white color tones, the particles are moving with low speed (minimal movement) and are form a circle in the center of the screen.

[0037] By issuing commands, in reaction to which the system generates pictorial representations, there is created a loop of exchange - there is established a dynamic in which the user "hypnotizes" the "artificial mind" (in a manner imitating coding), and the system hypnotizes them back toward cognitive disorientation - i.e. the system hypnotizes them back towards hypnotherapeutic "relaxation" (healing trance). In other words, the system's interaction comprises a feedback loop, where the user's voice commands induce states in the system, and the system's visual responses induce hypnotherapeutic effects in the user. In yet another words, the hypnotherapeutic interaction is created as a loop of exchange where the user "hypnotizes" the Al system with induction words, and the Al system responds with pictorial representations that induce a hypnotherapeutic state in the user.

[0038] The factors responsible for disorienting the user are semiotic inconsistency and representational obscureness. In other words, the system's visual responses hardly match the signification of the commands - for example, when the user says "sleep" (for calming down stage) the image "explodes" into a cascade of particles; example of such a pictorial representation is shown in Fig. 4. The pictorial representations which the system sends back are abstractions irrespective of the logic of human language - the pictorial representations produced by the Al system are abstractions that may not match the signification of the user commands, contributing to cognitive disorientation and hypnotherapeutic relaxation. Despite that, the resonance of interaction is strong enough to make the users speak of "communication on a deeper level". The two factors responsible for the therapeutic effect are (1) the abstract representation of the hypnotic state that enables the hypnotizing effect in a user - the system responds to the commands visually, using abstraction in place of language and logical thinking; (2) the individuation of the pictorial representation - the system reacts differently to different users responding to a user's specific voice and utterance.

[0039] The system responds only tothe selected, predefined word commands. The examples of visual representations semiotically inconsistent with and do not matching the semantic signification ("meaning") of the user's voice commands include:

[0040] Activation commands:

[0041] • "relax" (example effect: the particles explode in a palette of red, orange, yellow and white and scatter to the left with a delay producing a comet-like veil in the animation);

[0042] • "close your eyes" (example effect: the image morphs into the initial circle position to explode evenly in a green-blue-and-white combination of colours);

[0043] • "sink deeper" (see above);

[0044] • "go deeper" (see above);

[0045] • "imagine" (see above);

[0046] • "feel calm" (see above;)

[0047] • "surrender" (see above).

[0048] Calming down commands: • "do nothing" (example effect: the image may transform from a scatter to a cascade of particles and move down the screen changing colours from yellow-orange-white to white).

[0049] • "go to sleep" (see above)

[0050] • "release" (see above)

[0051] • "let go" (see above)

[0052] • "fading" (see above)

[0053] • "breathe in" (see above)

[0054] • "breathe out" (see above)

[0055] For each of the above-mentioned interactions (activation; calming down; waking up), the particles may exhibit the following movement: flow, cascade, shader, scatter, integrate, morph, transform, explode (simulate an explosion effect) - all based on and arranged around the initial "circle position" and randomized in terms of movement, scale and coIor each time the image changes to a command.

[0056] The behaviour of the produced image is different for each interaction with the user (and different for different users). Also, each follow-up image (reaction to subsequent commands in one user) depends on the shape of the previous image that the system decides to create.

[0057] The only consistent responses are the ones to activate the emergence from the "hypnotic state", i.e.:

[0058] • "emerge" (the particles return to the initial "circle position" and the image is still - there is no scattering, no movement);

[0059] • "open your eyes" (the particles return to the initial "circle position" and the image is still -there is no scattering, no movement).

[0060] Technologically, the system rests on two elements: the Python-fueled script with word recognition and the Unreal Engine for generating the particle imaging. The exemplary code is shown in Fig. 6.

[0061] A Python library called Porcupine was used for word recognition. Each hotword was pretrained using htt s: / / console. icovoice.ai / website and stored in hotwords directory. The detected words get counted for repetition detection and is sent to the Unreal engine.

[0062] The system generates abstract pixelated pictorial representations based on voice commands, using Unreal Engine and a Python library Porcupine. When a user issues a voice command, it is interpreted by voice recognition software. The textual command is then processed by the Porcupine library (https: / / github.com / Picovoice / porcupine), which uses algorithms to create pixelated images. These images are rendered using Unreal Engine.

[0063] The generated images have a unique conceptual relationship to the voice commands. They can either be the "opposite" of the command's semantic meaning, providing a contrasting visual representation, or bear a more random relation, offering an element of unpredictability. In both of these approaches, the pictorial representations do not match the signification of the user commands. This dual approach allows for diverse and intriguing visual outputs from simple voice inputs.

[0064] Below will be described the parameters defining the behavior of pictorial representation's pixels regarding color, speed, scattering, and other attributes.

[0065] Color parameters: The system relies on the Random Palette where colors are chosen randomly from a predefined set balancing between different colour tones to render the required effect of abstraction against a black background. The colours rely on different variance defined by a changing degree of variation from the base colour (usually ±20% hue variation). There is also a gradient effect applied across the image based on pixel position.

[0066] Pixel Movement Speed: The pixel movement oscillates between base and accelerated speed. The initial speed at which pixels are moved is approximately 5 pixels per second with a rate changing to increase speed over time based on varied acceleration (range from 1 pixel per second2to more).

[0067] Scattering Parameters: The pixels scatter directionally (radially) with a preferred direction bias of 30 to 70 % of bias. The base size of each pixel is 10x10 with variance range ±5 pixels). The opacity of pixels ranges from 0.8 to -0.05 per second.

[0068] Interaction with Voice Commands: The system relies on the 100% command influence which means that voice commands cause direct and immediate changes in the pixel behavior. There is a slight time delay of 0.5 seconds between receiving a command and its visual effect.

[0069] The pixels form the system specific patterns coming from a circular arrangement that scatters and reshapes based on a random distribution. The pattern transitions by sliding with not extra visual effect. Thus, porcupine Library handles the algorithmic generation of pixelated images based on the parameters defined above. It processes the input commands and translates them into pixel behavior rules.

[0070] Unreal Engine renders the generated images with high fidelity, ensuring smooth animations and transitions based on the defined parameters. It also manages the real-time interaction and visual feedback to the user.

[0071] The system has been trained to recognize and respond to the induction phrases used in the experiment and to create its own recognition content based on the feedback from each interaction. Interactions are recorded in the system so that the system learns from them and, based on that, is able to optimize its performance. The optimizing stage does not subserve any function or any specific expectation; rather, it develops to the system's "preference" or the system's own computational "logic."

[0072] As mentioned above, the system comprises a large, high-resolution screen for displaying pictorial representations, wired to a computer on which the system runs (for example a PC computer), a voice recognition program, and a high-fidelity microphone allowing the user to input voice commands. This setup is designed to provide an optimal user experience for voice command input and visual output, specifically targeting the accurate display of high-detail, pixelated images. Below the components of the system, according to an embodiment, will be described.

[0073] The PC (personal computer) may be a modern PC with a multi-core processor, at least 16GB of RAM, and a dedicated graphics card. The operating system may be for example Windows 10 / 11, macOS, or Linux.

[0074] For voice recognition, an advanced voice recognition software such as Dragon NaturallySpeaking, Google Speech-to-Text, or an open-source alternative like CMU Sphinx may be used. The voice recognition program integrates with the system's operating system and other software to process voice commands and control the display output. It allows for real-time speech processing, customizable command sets, and robust error handling. The system must accurately recognize and interpret user voice commands. Commands include actions like zooming, panning, switching images, and adjusting display settings. The microphone is preferably a high-fidelity, unidirectional microphone, with noise-cancelling capabilities, frequency response range of 20Hz to 20kHz, and USB or XLR connection. The microphone should be positioned at an optimal distance from the user to capture clear voice commands without interference. The microphone and voice recognition software should maintain high accuracy under different environmental conditions. The system should handle background noise and prevent false command triggers.

[0075] The display is preferably 27-inch or larger, with 4K (3840x2160 pixels) or higher resolution, to ensure crisp display of pixelated images - the high-resolution screen renders images accurately, highlighting the pixel-level details. Preferably it is OLED or LED display with high contrast ratio and wide color gamut. The display should refresh at a rate that prevents flickering or lag. This allows to display images, including highly detailed pixelated images, with clarity and precision.

[0076] The system preferably supports various image formats (JPEG, PNG, BMP, etc.). It should also respond to voice commands with minimal latency. The interface should be user-friendly, providing visual feedback on recognized commands and current system status. It should include touch or gesture control as a backup to voice commands. The architecture of the system should allow for easy upgrades of hardware components, such as the microphone or display, to keep pace with technological advancements. The system should be easy to set up and configure, with intuitive calibration for the microphone and voice recognition program. It should provide clear instructions and support for troubleshooting.

[0077] The system workflow, i.e. the systems' operation during use, is as follows.

[0078] The PC boots up and loads the operating system and voice recognition program. The microphone is initialized and calibrated for optimal voice capture. When the system is ready, the user speaks into the microphone. The voice recognition software processes the audio input, converting it into text commands. The commands are interpreted and executed by the system. Further, based on the voice commands, the system fetches and processes the required images. The images are displayed on the high-resolution screen, ensuring pixelperfect clarity. Further, the system provides visual and auditory feedback confirming the execution of commands. Continuous voice recognition ensures that subsequent commands are processed without interruption. This command recognition system architecture, with its high-fidelity microphone and large high-resolution screen, is designed to deliver a seamless user experience for both command input and visual output. The high-resolution display ensures that even pixelated images are rendered with exceptional clarity, making it suitable for applications that require detailed visual representation.

[0079] The system is also a research tool. When used in workshops, it extends through the use of the conceptual element that prepares the user for the experiment's experience and that subsequently reviews the experience outcome. During preparation, the users answer four "before use" questions:

[0080] 1. What is hypnosis? How do you understand the process in terms of its form, purpose and risks?

[0081] 2. What is Al? How do you understand its function, purpose and risks?

[0082] 3. What does the system (Hypnotic Al) imply? What does it mean?

[0083] 4. (Based on the set-up) What do you expect from the experiment / the experience?

[0084] During the review that follows the experience, the users answer four "after use" questions:

[0085] 1. Does your experience of the system (Hypnotic Al) match with your understanding of hypnosis? If yes, how so? If not, how so?

[0086] 2. Does your experience of Hypnotic Al match your understanding of Al? If yes, how so? If not, how so?

[0087] 3. What new possibility regarding Al comes from the experience of the system (Hypnotic Al)?

[0088] 4. How would you define Al based on your experience of the system (Hypnotic Al)?

[0089] The surveys synchronize the system's goals, which are to challenge the anthropomorphizing model of cognitive and affective computing. They also guide the users across the abstraction of the artificial mind to facilitate the experience of artificial intelligence as an alien intelligence interacting with us from a "vulnerable" state of non-logic, which rather than scare us may put us in awe. Feedback from the users— recorded in the system and from the survey— confirms that artistic methodologies and tools derived from the characteristics of a creative process create a perfect environment for human-AI interactions. This is because ABR ("art-based research framework") methods optimize the cultural charge of Al. They allow for less prejudiced and more reliable experimentation, one that is scientifically meaningful on the one hand and creatively heterogeneous on the other. Such an environment encourages unconstrained experimentation with human-AI interactions to expand our consciousness and overcome our prejudice.

[0090] The system simulates concretisation to rehearse the vulnerable moment of interacting outside of the code. It does it for both the human and the machine demystifying the nature of hypnosis and reclaiming the nature of an artificially intelligent system. Perhaps the biggest misconception about technology is that it is not part of nature. Just like the functional myth of technology being ominous in its mysterious complexity. The system advocates for the complexity of Al, which is not one but more-than-one cognitive possibility. It sees media practices in terms of interactive (and intra-active) encounters of human and non-human materials, wherein the human materials are the media subjects (users and their "organic element") and the non-human materials are the media objects and their environments.

Claims

Claims1. An interactive hypnotherapeutic computer-implemented system based on artificial intelligence, comprising: a microphone for receiving voice commands from a user; a display device for displaying dynamic pictorial representations generated by the system; a computing device operatively connected to the display device and the microphone, characterised in that the computing device is configured to execute a computer program implementing a hypnotherapeutic interaction with the user based on received voice commands, whereby the computing device comprises: a word recognition module configured to recognize predefined induction words or phrases from the received voice commands; an artificial intelligence, Al, response generation module trained to respond to the recognized induction words or phrases by generating corresponding dynamic pictorial representations, created by moving particles, on the display device, and configured to record interactions and optimize responses based on user-specific feedback; wherein the system transitions through three stages represented by distinct visual patterns of dynamic pictorial representations, induced by the predefined induction commands, wherein the dynamic pictorial representations vary in parameters defining the behavior of pictorial representation's particles, in particular vary in color and / or speed and / or movement pattern and / or scattering of representations' particles, and wherein the visual representations generated by the system are semiotically inconsistent and do not match the semantic signification of the user's voice commands.

2. The system according to claim 1, characterized in that the system transitions through stages of activation, calming down and waking up, wherein the pictorial representation of the activation stage is represented by warm color tones, highspeed particle movement, and high particle scattering;the pictorial representation of the calming down stage is represented by cold color tones, low- speed particle movement, and low particle scattering with concentrated particle formation in the center of the screen; the pictorial representation of the waking up stage is represented by wite color tone, minimal particle movement, and no particle scattering with particles forming a circle in the center of the screen.

3. The system according to claim 2, characterized in that the activation stage is induced by induction commands selected from a group including: "relax", "close your eyes", "sink deeper", "go deeper", "imagine", "feel calm", and "surrender".

4. The system according to claim 2 or 3, characterized in that the calming down stage is induced by induction commands selected from a group including: "do nothing", "go to sleep", "release", "let go", "fading", "breathe in", and "breathe out".

5. The system according to any of the preceding claims, characterized in that the waking up stage is induced by induction commands selected from a group including: "emerge", and "open your eyes".

6. The system according to any of the preceding claims, characterized in that the artificial intelligence module is configured to record interactions with users and optimizing its responses based on feedback from these interactions, developing unique pictorial representations and responses for different users.

7. The system according to claim 6, characterized in that the optimization is developed according to the system's own computational logic.

8. The system according to any of the preceding claims, characterized in that the Al response generation module is anchored in software implemented in Python script for word recognition and uses the Unreal Engine for generating pictorial representations.

9. The system according to any of the preceding claims, characterized in that the Al response generation module uses open-end programming to ensure various protocols and modes that prevent the domination of preliminary coding.

10. A computer program product, stored on a non-transitory computer-readable medium, comprising instructions that, when executed by a computer system, cause the computer system to: a. recognize specific induction words or phrases from voice commands received via a microphone; b. in response to the recognized induction words or phrases, generate corresponding dynamic pictorial representations, comprising particles, on a display device using an artificial intelligence response generation module, the pictorial representations being distinct for three stages of the system and varying in color, speed, movement pattern and scattering of representations' particles; c. record interactions and optimize responses based on user-specific feedback; wherein the visual representations are semiotically inconsistent and do not match the semantic signification of the user's voice commands.

Images