Privacy protection type proof of creation process, device, equipment and medium

Abstract

The application discloses a privacy protection type right proof method, device and equipment of a creation process and a medium, relates to the field of copyright right, and comprises the following steps: obtaining a creation process event, dividing the creation process event into a creation event fragment based on a preset fragment division condition, generating a privacy classification result by classifying the creation event fragment according to a preset classification rule; generating a local abstract commitment based on a local content abstract and a version abstract according to the privacy classification result; converting the creation event fragment into a chain node, connecting each chain node into a creation continuity chain based on the version evolution relationship in the creation process and the time sequence of the creation event fragment; determining the creation continuity credibility according to the creation continuity chain, and determining the AI auxiliary call boundary credibility according to the AI auxiliary call boundary abstract; and generating a privacy protection process proof according to a preset privacy proof strategy to output a creation process proof package. In this way, the right can be proved under the premise of protecting the creation privacy of the original manuscript.

Description

Technical Field

[0001] This invention relates to the field of copyright confirmation, and in particular to a method, apparatus, equipment, and medium for privacy-protecting copyright confirmation during the creation process. Background Technology

[0002] Existing digital work ownership verification schemes typically use the final draft, work hash, timestamp, or creation log as evidence. For creative processes requiring continuous editing, such as writing, painting, editing, and design, these schemes present two conflicting problems: On the one hand, submitting only the final draft hash or registering the timestamp makes it difficult to prove that the work underwent a continuous human creative process, and also makes it difficult to distinguish between human creation, AI-assisted creation, and purely automatically generated content; on the other hand, directly submitting complete drafts, pen strokes, keyboard rhythms, material versions, AI call logs, and creation environment data exposes unpublished content, creator's behavioral habits, device information, creative strategies, and sources of commercial materials.

[0003] Currently, existing digital copyright verification methods typically employ technologies such as work hashing, trusted timestamps, blockchain notarization, electronic signatures, content certificate lists, log auditing, or AI-generated content identifiers. These can prove the existence of a work file or metadata at a specific time, but they are insufficient to prove how the work was created.

[0004] Closer solutions include: the first type is a blockchain-based copyright registration or NFT ownership verification scheme, which writes the work summary, author identity summary, and transaction records on the chain; the second type is a content source certificate scheme, which records the source of the work, editing history, and signature information through a content certificate list; the third type is a process proof scheme, which proves the creation process by recording process events such as keyboard input, clipboard activity, processes, AI tool calls, and time intervals; and the fourth type is a zero-knowledge process proof scheme, which attempts to prove the validity of certain creative attributes without disclosing complete process data.

[0005] In conclusion, how to establish ownership rights while protecting the privacy of original creators is an urgent issue that needs to be addressed. Summary of the Invention

[0006] In view of this, the purpose of this invention is to provide a privacy-protecting method, apparatus, device, and medium for rights confirmation during the creative process, which can confirm rights while protecting the privacy of the original manuscript. The specific solution is as follows: Firstly, this application provides a privacy-protected method for proving rights during the creation process, including: The creation process events generated during the creation of the target work are obtained, and the creation process events are divided into various creation event segments based on preset segmentation conditions. The creation event segments are then classified for privacy to generate corresponding privacy classification results. A partial summary commitment is generated based on the privacy classification result for the creation event fragment; the partial summary commitment is used to characterize the version evolution relationship of the creation event fragment without disclosing the detailed data of the creation event fragment. Each of the creation event fragments is transformed into chain nodes, and based on version evolution relationships and / or time order, each of the chain nodes is connected into a creation continuity chain; wherein, the chain nodes in the creation continuity chain are bound to the local summary commitment, time anchor point and AI-assisted call boundary summary of the corresponding creation event fragment, and the AI-assisted call boundary summary is a privacy representation of the relationship between AI-assisted call and manual modification in the corresponding creation event fragment; The credibility of the creative continuity of the target work is determined based on the creative continuity chain, and the credibility of the corresponding AI-assisted invocation boundary is determined based on the AI-assisted invocation boundary summary. Based on the preset privacy proof strategy corresponding to the target rights confirmation scenario, a privacy protection process proof is generated based on the creation continuity chain, the creation continuity credibility, the AI-assisted call boundary credibility, and the preset credibility threshold, and a creation process proof package containing the privacy protection process proof is output.

[0007] Optionally, the acquisition of creative process events generated during the creation of the target work involves dividing these events into various creative event segments based on preset segmentation conditions, including: During the creation process of the target work, events of the creation process are collected in real time and written to the local event cache; In the local event cache, if the collected creation process events meet any of the triggering conditions in the preset segmentation conditions, then the continuous creation process events that meet the triggering conditions are extracted to construct the corresponding creation event segments; wherein, the triggering conditions include any one or more of the following: reaching a preset time window, reaching a preset number of key operations, reaching a version node, and a work area switch.

[0008] Optionally, the privacy classification results include public level, verified level, and confidential level; The process of performing privacy classification on the creation event fragments to generate corresponding privacy classification results includes: The sensitivity level of each field contained in the creative event fragment is determined according to the preset sensitivity level classification conditions, and the sensitivity level corresponding to each field is generated. The sensitivity level that meets the highest preset level condition among all fields is determined as the privacy classification result of the creation event fragment; Accordingly, the commitment to generate a partial summary of the creation event fragment based on the privacy classification result includes: Generate a displayable summary for the fields whose sensitivity level is public. Based on preset commitment generation rules, for fields with the sensitivity level of the verification level, a corresponding partial content summary and proof input for generating proof of the privacy protection process are generated; Fields with the sensitivity level set to the confidentiality level are only encrypted and backed up locally.

[0009] Optionally, the step of generating a corresponding partial content digest and proof input for generating a privacy protection process proof for the field with the sensitivity level of the verification level based on the preset commitment generation rule includes: The pre-version summary, post-version summary, and local difference summary of the fields at the verification level are used to perform summary calculation operations to generate a summary with invisible content; the pre-version summary is the summary of the version before fragment processing, and the post-version summary is the summary of the version after fragment processing. The event type distribution, human input density, version saving times, material import times, and AI-assisted call types of the fields at the verification level are summarized to generate a behavioral statistical summary. Based on the content-invisible summary and the behavior statistics summary, a corresponding local content summary is determined; The proof input is generated by combining the privacy classification result, quality marker, and AI-assisted call boundary summary of the fields of the verification level.

[0010] Optionally, determining the credibility of the creative continuity of the target work based on the creative continuity chain includes: A segment gap penalty term is determined based on the portion of the time interval between adjacent creative event segments that exceeds the preset maximum normal segment interval; wherein, the segment gap penalty term is used to reduce the credibility of creative event segments with abnormal time gaps; An automatic input anomaly penalty term is determined based on the similarity between the input rhythm vector and the preset automatic input rhythm vector, and the coefficient of variation of the input rhythm vector; wherein, the automatic input anomaly penalty term is used to reduce the credibility of the creative continuity when the similarity is higher than a similarity threshold and the coefficient of variation is lower than a variation threshold. The credibility of the creative continuity is determined by the first preset credibility calculation formula, based on the artificial event density of each creative event segment, the time interval between adjacent creative event segments, the segment gap penalty item, and the automatic input anomaly penalty item. The first preset credibility calculation formula is: ; in, This indicates the credibility of the stated creative continuity. This represents the Sigmoid normalization function, where n represents the number of creation event fragments. This represents the effective artificial event density of the i-th creative event fragment. This indicates the time interval between adjacent creative event segments. Indicates the average of time intervals. This indicates a fragment gap penalty term. This indicates that the penalty for errors will be automatically entered.

[0011] Optionally, determining the corresponding AI-assisted invocation boundary confidence level based on the AI-assisted invocation boundary summary includes: The credibility of the AI-assisted call boundary is determined by the second preset credibility calculation formula based on the difference between the amount of manual modification, the amount of AI output, the distribution of AI output summary and the distribution of manually modified summary in the AI-assisted call boundary summary. The second preset credibility calculation formula is: ; in, This indicates the credibility of the AI-assisted invocation boundary. This represents the amount of effective modifications made by humans after the AI ​​output. Indicates the amount of AI output. This indicates a smoothing term to prevent the denominator from being zero. Indicates the Kullback-Leibler divergence. This indicates the distribution of AI output summaries. This indicates the distribution of the abstracts after manual modification. This indicates the proportion of large, unexplained output segments.

[0012] Optionally, the step of generating a privacy protection process proof based on the preset privacy proof strategy corresponding to the target rights confirmation scenario, the creation continuity chain, the creation continuity credibility, the AI-assisted call boundary credibility, and a preset credibility threshold, and outputting a creation process proof package containing the privacy protection process proof, includes: The categories of data that are allowed to be disclosed and the corresponding granularity of data disclosure are determined based on the target rights confirmation scenarios; wherein, the target rights confirmation scenarios include any one or more of the following: rights confirmation and registration scenarios, platform review scenarios, ownership dispute scenarios, and infringement evidence scenarios; Based on the data category and the data disclosure granularity, corresponding privacy protection process proofs are extracted from the local summary commitment, the creation continuity chain, the AI-assisted call boundary summary, and the time anchor; wherein, the privacy protection process proof is used to demonstrate that each chain node is generated and valid by the corresponding creation event fragment without disclosing the detailed data of the creation event fragment; Determine whether the credibility of the creation continuity is greater than or equal to a first preset credibility threshold, and whether the credibility of the AI-assisted call boundary is greater than or equal to a second preset credibility threshold, so as to obtain the corresponding credibility determination result; Based on the extracted privacy protection process proof and the credibility determination result, a creation process proof package corresponding to the target work is generated.

[0013] Secondly, this application provides a privacy-protected rights confirmation device for the creation process, comprising: The result generation module is used to acquire the creation process events generated during the creation of the target work, divide the creation process events into various creation event segments based on preset segment division conditions, and perform privacy classification on the creation event segments to generate corresponding privacy classification results. The commitment generation module is used to generate a partial summary commitment of the creation event fragment based on the privacy classification result; the partial summary commitment is used to characterize the version evolution relationship of the creation event fragment without disclosing the detailed data of the creation event fragment. The node connection module is used to convert each of the creation event fragments into chain nodes, and connect each of the chain nodes into a creation continuity chain based on version evolution relationship and / or time order; wherein, the chain nodes in the creation continuity chain are bound to the local summary commitment, time anchor point and AI-assisted call boundary summary of the corresponding creation event fragment, and the AI-assisted call boundary summary is a privacy representation of the relationship between AI-assisted call and manual modification in the corresponding creation event fragment; The credibility generation module is used to determine the credibility of the creation continuity of the target work based on the creation continuity chain, and to determine the corresponding credibility of the AI-assisted call boundary based on the AI-assisted call boundary summary. The proof package generation module is used to generate a privacy protection process proof based on the preset privacy proof strategy corresponding to the target rights confirmation scenario, the creation continuity chain, the creation continuity credibility, the AI-assisted call boundary credibility, and the preset credibility threshold, and output a creation process proof package containing the privacy protection process proof.

[0014] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement a privacy-preserving method for proving ownership of rights during the creation process as described above.

[0015] Fourthly, this application provides a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements a privacy-protected method for proving ownership of rights as described above in the creation process.

[0016] In summary, this application obtains creative process events generated during the creation of the target work, divides these events into fragments based on preset fragment segmentation conditions, and performs privacy grading on these fragments to generate corresponding privacy grading results. It then generates partial summary commitments for each fragment based on the privacy grading results. These partial summary commitments characterize the version evolution relationship of each fragment without disclosing detailed data. Each fragment is transformed into a chain node, and based on version evolution relationships and / or chronological order, these chain nodes are connected to form a creative continuity chain. Each chain node in the creative continuity chain is bound to a corresponding... The creation event fragment includes a local summary commitment, a time anchor, and an AI-assisted call boundary summary, where the AI-assisted call boundary summary is a privacy-enhanced representation of the relationship between AI-assisted calls and manual modifications in the corresponding creation event fragment. The creation continuity credibility of the target work is determined based on the creation continuity chain, and the corresponding AI-assisted call boundary credibility is determined based on the AI-assisted call boundary summary. According to a preset privacy proof strategy corresponding to the target rights confirmation scenario, a privacy-protected process proof is generated based on the creation continuity chain, the creation continuity credibility, the AI-assisted call boundary credibility, and a preset credibility threshold. A creation process proof package containing the privacy-protected process proof is then output.

[0017] As shown above, this application first obtains various events generated during the creation process of the target work, and divides them into several creation event fragments according to preset fragment segmentation conditions. Then, it performs privacy classification on these fragments according to the classification rules to obtain the corresponding privacy classification results. On this basis, for each creation event fragment, a local summary commitment is generated using its local content summary and version summary. Next, each creation event fragment is transformed into a chain node, and these nodes are connected into a creation continuity chain according to the version evolution relationship and time sequence during the creation process. Based on the creation continuity chain, the credibility of the creation continuity of the target work during the creation process can be calculated, and the credibility of the AI-assisted call boundary can be determined based on the AI-assisted call situation. Finally, according to the preset privacy proof strategy corresponding to the target rights confirmation scenario, the privacy protection process proof is generated by comprehensively considering the local summary commitment, the creation continuity chain, the AI-assisted call boundary summary, the time anchor, the above two credibility values, and the preset credibility threshold, and is encapsulated in the creation process proof package for output. In this way, without exposing the detailed data of the creation event fragments, it can show the verification party that each chain node is indeed generated by the corresponding creation event fragment and is valid. At the same time, it proves that the credibility of the creation continuity and the credibility of the AI-assisted call boundary both meet the preset threshold requirements, thus providing reliable privacy evidence for the authenticity and compliance of the work creation process. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0019] Figure 1 This application discloses a flowchart of a privacy-protected method for proving ownership during the creation process. Figure 2 This is a schematic diagram of a creative continuity chain data structure disclosed in this application; Figure 3 This application discloses a sequence diagram for the generation and verification of a proof package for a human creative process. Figure 4 This is a schematic diagram of a privacy-protecting rights confirmation device for the creative process disclosed in this application; Figure 5 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Currently, existing digital copyright confirmation methods typically employ technologies such as work hashing, trusted timestamps, blockchain notarization, electronic signatures, content certificate lists, log auditing, or AI-generated content identifiers. These can prove the existence of a work file or metadata at a specific time, but are insufficient to prove how the work was created. Closer solutions include: the first type is blockchain-based copyright registration or NFT confirmation schemes, which write work summaries, author identity summaries, and transaction records onto the chain; the second type is content source certificate schemes, which record the work's source, editing history, and signature information through a content certificate list; the third type is process proof schemes, which prove the creation process by recording process events such as keyboard input, clipboard activity, processes, AI tool calls, and time intervals; and the fourth type is zero-knowledge process proof schemes, which attempt to prove certain creative attributes without disclosing complete process data. To address the aforementioned technical problems, this application discloses a privacy-protected copyright confirmation method, apparatus, device, and medium for the creation process, enabling copyright confirmation while protecting the privacy of the original manuscript.

[0022] See Figure 1 As shown, this invention discloses a privacy-protected method for proving ownership during the creation process, including: Step S11: Obtain the creation process events generated during the creation of the target work, divide the creation process events into various creation event segments based on preset segment division conditions, and perform privacy classification on the creation event segments to generate corresponding privacy classification results.

[0023] In this embodiment, during the creation of the target work, creation process events are collected in real time and written to a local event cache. If any of the collected creation process events in the local event cache meets any of the triggering conditions in the preset segmentation conditions, then continuous creation process events that meet the triggering conditions are extracted to construct corresponding creation event segments. The triggering conditions include reaching a preset time window, reaching a preset number of key operations, reaching a version node, and a work area switch. Specifically, the creation client collects creation process events in writing, drawing, editing, design, or multimedia editing software. These creation process events include text addition, deletion, modification, pen strokes, layer editing, material import, clip adjustment, keyboard input rhythm, version saving, AI-assisted calls, export operations, and device trust status. The original work content is not directly uploaded during collection; instead, an event cache is formed locally. Based on the time window, operation type, work area, version node, and material object, continuous events are divided into multiple creation event segments. A creation event segment can be represented as: ; in, This represents the i-th creation event fragment; Indicates the segment number; and These represent the start time and the end time, respectively. It represents a set of event types, with a preset type table, including text addition, deletion and modification, pen stroke trajectory, layer editing, material import, clip adjustment, keyboard input rhythm, version saving, AI-assisted calling, export operation, device trust status, etc. It can be expanded according to software category. During collection, each original event is mapped to a type label according to the operation semantics. This represents a summary of the version before fragment processing; This represents a summary of the processed version. Indicates a partial summary commitment; Indicates privacy classification; This indicates the quality of the fragment. It should be noted that this embodiment does not use a complete creation log as the subject of external disclosure, but rather uses fragments of the creation event as the smallest unit of proof.

[0024] Furthermore, each creative event segment is classified into privacy levels, including public, verified, and confidential levels. This privacy classification process involves: determining the sensitivity level of each field within the creative event segment based on preset sensitivity level criteria, generating a sensitivity level for each field; and determining the sensitivity level of the field that meets the highest preset level criteria as the privacy classification result for the creative event segment. Specifically, when classifying the privacy of each creative event segment, a refined sensitivity assessment is implemented at the level of each field within the segment. Based on preset sensitivity level criteria, the sensitivity level of each field within the segment is determined one by one. These criteria may involve whether the field contains personal identification information, sensitive operation records, or private creative content, thereby generating a corresponding sensitivity level for each field, such as public, verified, or confidential. Subsequently, the sensitivity level determination results of all fields under the segment are summarized, and the levels of all fields are compared. The sensitivity level with the strictest protection requirements is selected as the final privacy classification result for the creative event segment. For example, if most fields in a data segment are classified as public, but even one field is classified as confidential, the entire segment will be classified as confidential. This aggregation mechanism, based on the highest sensitivity level, ensures that no single highly sensitive field will be downgraded due to the presence of other less sensitive fields, thus maximizing the protection of user privacy and providing a solid and reliable classification basis for subsequent data access control and de-identification processes.

[0025] Step S12: Generate a partial summary commitment for the creation event fragment based on the privacy classification result.

[0026] In this embodiment, a displayable summary is generated for fields with a sensitivity level of "public". Based on preset commitment generation rules, a corresponding partial content summary and proof input for generating a privacy protection process proof are generated for fields with a sensitivity level of "verification". Fields with a sensitivity level of "confidential" are only encrypted and backed up locally. Specifically, summary calculation operations are performed on the pre-version summary, post-version summary, and partial difference summary of the verification-level fields to generate a content-invisible summary; the pre-version summary is the summary of the version before fragment processing, and the post-version summary is the summary of the version after fragment processing. Summary calculation operations are performed on the event type distribution, manual input density, version saving times, material import times, and AI-assisted call types of the verification-level fields to generate a behavioral statistics summary. A corresponding partial content summary is determined based on the content-invisible summary and the behavioral statistics summary. The privacy classification result, quality marker, and AI-assisted call boundary summary of the verification-level fields are combined to generate the proof input. Specifically, differentiated summary and protection operations are performed on fields with different sensitivity levels. For public-level fields, a displayable summary is directly generated, and its content can be directly viewed and retrieved. The processing of verification-level fields follows preset commitment generation rules, producing both a partial content summary and proof input for generating privacy-protected process proofs. First, the pre-version summary, post-version summary, and partial difference summary of the field are aggregated to obtain a content-invisible summary, providing an anchor point for integrity verification while hiding the original content. Then, behavioral metadata associated with the field, such as event type distribution, human input density, version save count, material import count, and AI-assisted call type, are summarized to form a behavioral statistical summary. Subsequently, based on the content-invisible summary and the behavioral statistical summary, the partial content summary of the field is derived. Simultaneously, the privacy classification results, quality markers, and AI-assisted call boundary summaries of the verification-level field are combined to form the proof input. Through these three layers of summary processing, the verifier can only see the existence of fragments, fragment order, version evolution relationships, and proof results that meet the thresholds; they cannot deduce the original content, complete creation habits, or AI prompts. For confidential fields, the strictest protection is implemented, with encrypted backups performed only locally, generating no external summaries, thus preventing information leakage at the source.

[0027] It should be noted that the partial summary commitment is used to characterize the existence, sequence, and version evolution of the creation event fragments without disclosing detailed data of the creation event fragments.

[0028] Step S13: Convert each of the creation event fragments into chain nodes, and connect each of the chain nodes into a creation continuity chain based on version evolution relationship and / or time order.

[0029] In this embodiment, multiple creative event fragments are connected according to version evolution relationships and chronological order as follows: Figure 2 The creative continuity chain is shown. The creative continuity chain can be represented as: ; Each node At least include fragment commitments Precursor node summary Time anchor point Stage tags AI-assisted call boundary summary Device Trusted Status Summary .

[0030] Furthermore, generate chain node commitments for each chain node. : ; in, This represents the commitment of the i-th chain node. `||` represents a hash or commitment function, and `||` represents a concatenation operation. Chain node commitment. This is not a simple hash of the final draft of the work, but rather a hash that simultaneously binds fragment commitments, preceding nodes, time anchors, AI-assisted call boundaries, and device trust status. If an attacker subsequently adds fragments, changes their order, or deletes intermediate fragments, the commitments of subsequent chain nodes will change accordingly, thus increasing the difficulty of forging a continuous creation process. Chain nodes are linked through version summaries, local difference summaries, and time interval constraints. The AI-assisted call boundary summary is a privacy-preserving representation of the relationship between AI-assisted calls and human modifications within the corresponding creation event fragment.

[0031] Step S14: Determine the credibility of the creative continuity of the target work based on the creative continuity chain, and determine the corresponding credibility of the AI-assisted call boundary based on the AI-assisted call boundary summary.

[0032] In this embodiment, a segment gap penalty is determined based on the portion of the time interval between adjacent creative event segments that exceeds a preset maximum normal segment interval; wherein, the segment gap penalty is used to reduce the credibility of creative event segments with abnormal time gaps; an automatic input anomaly penalty is determined based on the similarity between the input rhythm vector and a preset automatic input rhythm vector, and the coefficient of variation of the input rhythm vector; wherein, the automatic input anomaly penalty is used to reduce the credibility of the creative continuity when the similarity is higher than a similarity threshold and the coefficient of variation is lower than a variation threshold; the credibility of the creative continuity is determined based on the artificial event density of each creative event segment, the time interval between adjacent creative event segments, the segment gap penalty, and the automatic input anomaly penalty using a first preset credibility calculation formula. Specifically, to avoid falsifying the creative trajectory afterward, time gaps, abnormal pasting, and automatic input in the creative continuity chain are specifically detected. Segment gap penalty It can be represented as: ; in, Indicates the fragment gap penalty item This indicates the maximum allowed normal segment interval. When the time interval between multiple adjacent creative event segments exceeds [a certain limit], [the limit is set to 1]. hour It approaches 1 as the excess increases; used to reduce the credibility of centralized editing after filling logs or long periods of blankness.

[0033] Automatically input exception penalty items It can be represented as: ; in, Indicates automatic input of abnormal penalty items Represents the j-th input rhythm vector This indicates preset macro playback or automatic input of rhythm vectors; Sim represents the similarity function. The coefficient of variation of the j-th input rhythm vector represents the rhythm variation threshold; I represents the indicator function. When the input rhythm simultaneously satisfies both high macro playback similarity and low coefficient of variation, it is identified as an automatic input anomaly.

[0034] It should be noted that the credibility of creative continuity Instead of simple weighted summation, a composite calculation model is used, incorporating event density logarithmic aggregation, time interval stability penalty, fragment gap penalty, and automatic input anomaly penalty. The credibility calculation formula is as follows: ; in, Indicates the credibility of the creative continuity. This represents the Sigmoid normalization function; n represents the number of creation event fragments. This represents the effective artificial event density of the i-th creation event fragment. Indicates the time interval between adjacent creative event segments. Mean of time interval Indicates the fragment gap penalty item This indicates an automatic input anomaly penalty item. It should be noted that the actual creative process typically has a stable but not completely uniform event density and time interval; if logs are subsequently added in batches, abnormal uniformity, abnormal density, large gaps, or automatic input characteristics often appear. By taking the logarithm of the event density, we avoid individually high-density segments unreasonably inflating the overall credibility; by using the time interval standard deviation, we identify abnormally concentrated or abnormally blank creative processes; and by... and The system will convert log supplementation, macro playback, large-scale pasting, and automatic input modes into credibility penalties.

[0035] Meanwhile, the AI-assisted invocation boundary confidence level Sa is used to prove an interpretable relationship between the AI ​​tool invocation and subsequent manual modification, rather than simply recording whether AI was invoked. This application incorporates the ratio of manual editing after AI output, the distribution difference between AI output content and manually modified content, and the proportion of unexplained large sections of output into the calculation, i.e., it is determined through the second preset confidence level calculation formula, which is: ; in, Indicates the credibility of the AI-assisted invocation boundary. This represents the amount of effective modifications made by humans after the AI ​​output. Indicates AI output This represents a smoothing term to prevent the denominator from being zero. Indicates Kullback-Leibler divergence Represents the distribution of AI output summaries Distribution of manually modified abstracts This indicates the proportion of large, unexplained output segments.

[0036] Step S15: Based on the preset privacy proof strategy corresponding to the target rights confirmation scenario, and based on the creation continuity chain, the creation continuity credibility, the AI-assisted call boundary credibility, and the preset credibility threshold, generate a privacy protection process proof, and output a creation process proof package containing the privacy protection process proof.

[0037] In this embodiment, the data categories that can be disclosed and the corresponding data disclosure granularity are determined according to the target rights confirmation scenario. The target rights confirmation scenario includes any one or more of the following: rights registration scenario, platform review scenario, ownership dispute scenario, and infringement evidence scenario. Based on the data categories and the data disclosure granularity, corresponding privacy protection process proofs are extracted from the local summary commitment, the creative continuity chain, the AI-assisted call boundary summary, and the time anchor. It is determined whether the credibility of the creative continuity is greater than or equal to a first preset credibility threshold, and whether the credibility of the AI-assisted call boundary is greater than or equal to a second preset credibility threshold, to obtain the corresponding credibility judgment result. Based on the extracted privacy protection process proofs and the credibility judgment result, a corresponding creative process proof package for the target work is generated.

[0038] Specifically, privacy proof strategies can determine different disclosure granularities based on the verification scenario, as shown in Table 1: Table 1

[0039] Based on the selective generation of privacy-preserving process proofs in the verification scenario, these proofs do not directly prove the content of the work itself, nor do they disclose the complete creation process events. Instead, they prove that the following propositions are simultaneously true: First, the chain nodes in the creation continuity chain promise C. i All are generated from corresponding creative event fragments, and the summaries of the preceding and following nodes are continuous; secondly, the credibility of creative continuity Sc is not lower than the continuity threshold θ. c Third, the credibility S of AI-assisted call boundaries. a Not lower than the AI-assisted call boundary threshold θ a Fourth, the work summary publicly disclosed to the verifier is consistent with the final draft summary in the blockchain; fifth, the original manuscript, prompts, pen strokes, and keyboard rhythms are only used as private inputs or local data in the commitment and proof generation, and are not disclosed to the verifier as public inputs. Through the above propositional arrangement, this application uses zero-knowledge proofs for process attribute verification, rather than generalizing to prove the validity of certain data, thus avoiding ordinary blockchain data zero-knowledge verification schemes while enabling the verifier to obtain sufficient technical confirmation. Public Declaration of Zero-Knowledge Process Proof It can be represented as: ; in, Public statement indicating proof of privacy protection process "valid" indicates that the chain node's commitment is valid. Represents the final summary in the chain This indicates a publicly available abstract of the work that is visible to the verifier. The verifier then... The process attributes are confirmed, but the original manuscript, complete log, complete prompts, or complete device behavior profile will not be obtained. Finally, all necessary materials are packaged into a creation process verification package, which the creator can send to the verifier to prove the creation process.

[0040] As shown above, this application first obtains various events generated during the creation process of the target work, and divides them into several creation event fragments according to preset fragment segmentation conditions. Then, it performs privacy classification on these fragments according to the classification rules to obtain the corresponding privacy classification results. On this basis, for each creation event fragment, a local summary commitment is generated using its local content summary and version summary. Next, each creation event fragment is transformed into a chain node, and these nodes are connected into a creation continuity chain according to the version evolution relationship and time sequence during the creation process. Based on the creation continuity chain, the credibility of the creation continuity of the target work during the creation process can be calculated, and the credibility of the AI-assisted call boundary can be determined based on the AI-assisted call situation. Finally, according to the preset privacy proof strategy corresponding to the target rights confirmation scenario, the privacy protection process proof is generated by comprehensively considering the local summary commitment, the creation continuity chain, the AI-assisted call boundary summary, the time anchor, the above two credibility values, and the preset credibility threshold, and is encapsulated in the creation process proof package for output. In this way, without exposing the detailed data of the creation event fragments, it can show the verification party that each chain node is indeed generated by the corresponding creation event fragment and is valid. At the same time, it proves that the credibility of the creation continuity and the credibility of the AI-assisted call boundary both meet the preset threshold requirements, thus providing reliable privacy evidence for the authenticity and compliance of the work creation process.

[0041] As can be seen from the previous embodiment, this application discloses a privacy-protected method for proving ownership during the creation process, which can confirm ownership while protecting the privacy of the original artwork. Next, taking a student's creation of a digital illustration as an example, regarding... Figure 3 The privacy-protected method for proving ownership of the creation process is explained in detail.

[0042] The system collects events related to students' brushstrokes, layer additions, color adjustments, material imports, and version saves within the drawing software. A creative event segment is generated every 5 minutes or every 50 key operations, along with a partial summary commitment for each segment. Students may use AI tools to generate compositional references mid-process; the AI ​​call time, call type, input summary, output summary, and the proportion of areas subsequently manually redrawn are recorded, but the complete prompts and full AI output are not saved.

[0043] Once the artwork is completed, a creative continuity chain containing 32 creative event segments is generated, and a summary of each stage chain is written to the consortium blockchain or a trusted timestamp service. During platform review, students only submit a proof package of the human creative process. The verifier can verify that the artwork has a continuous evolution process from sketch, line drawing, coloring to the final draft, and can verify that the intensity of human modifications after AI-assisted invocation exceeds a threshold, without needing to view the complete draft and complete brushstroke behavior.

[0044] See Figure 4 As shown, this embodiment of the invention discloses a privacy-protected rights verification device for the creation process, which may specifically include: The result generation module 11 is used to acquire the creation process events generated during the creation of the target work, divide the creation process events into various creation event segments based on preset segment division conditions, and perform privacy classification on the creation event segments to generate corresponding privacy classification results. The commitment generation module 12 is used to generate a partial summary commitment of the creation event fragment based on the privacy classification result; the partial summary commitment is used to characterize the version evolution relationship of the creation event fragment without disclosing the detailed data of the creation event fragment. The node connection module 13 is used to convert each of the creation event fragments into chain nodes, and connect each of the chain nodes into a creation continuity chain based on version evolution relationship and / or time order; wherein, the chain nodes in the creation continuity chain are bound to the local summary commitment, time anchor point and AI-assisted call boundary summary of the corresponding creation event fragment, and the AI-assisted call boundary summary is a privacy representation of the relationship between AI-assisted call and manual modification in the corresponding creation event fragment; Credibility generation module 14 is used to determine the credibility of the creation continuity of the target work based on the creation continuity chain, and to determine the corresponding credibility of the AI-assisted call boundary based on the AI-assisted call boundary summary. The proof package generation module 15 is used to generate a privacy protection process proof based on the preset privacy proof strategy corresponding to the target rights confirmation scenario, the creation continuity chain, the creation continuity credibility, the AI-assisted call boundary credibility, and the preset credibility threshold, and output a creation process proof package containing the privacy protection process proof.

[0045] As shown above, this application first obtains various events generated during the creation process of the target work, and divides them into several creation event fragments according to preset fragment segmentation conditions. Then, it performs privacy classification on these fragments according to the classification rules to obtain the corresponding privacy classification results. On this basis, for each creation event fragment, a local summary commitment is generated using its local content summary and version summary. Next, each creation event fragment is transformed into a chain node, and these nodes are connected into a creation continuity chain according to the version evolution relationship and time sequence during the creation process. Based on the creation continuity chain, the credibility of the creation continuity of the target work during the creation process can be calculated, and the credibility of the AI-assisted call boundary can be determined based on the AI-assisted call situation. Finally, according to the preset privacy proof strategy corresponding to the target rights confirmation scenario, the privacy protection process proof is generated by comprehensively considering the local summary commitment, the creation continuity chain, the AI-assisted call boundary summary, the time anchor, the above two credibility values, and the preset credibility threshold, and is encapsulated in the creation process proof package for output. In this way, without exposing the detailed data of the creation event fragments, it can show the verification party that each chain node is indeed generated by the corresponding creation event fragment and is valid. At the same time, it proves that the credibility of the creation continuity and the credibility of the AI-assisted call boundary both meet the preset threshold requirements, thus providing reliable privacy evidence for the authenticity and compliance of the work creation process.

[0046] In some specific implementations, the result generation module 11 may specifically include: An event caching unit is used to collect creation process events in real time and write them to a local event cache during the creation of the target work. The fragment construction unit is used to extract continuous creative process events that meet the triggering conditions in the local event cache if the collected creative process events meet any of the preset fragment division conditions, and construct corresponding creative event fragments; wherein, the triggering conditions include any one or more of the following: reaching a preset time window, reaching a preset number of key operations, reaching a version node, and a work area switching.

[0047] In some specific implementations, the privacy classification results include public level, verification level, and confidential level; the result generation module 11 may specifically include: The sensitivity level generation unit is used to determine the sensitivity level of each field contained in the creative event fragment according to the preset sensitivity level division conditions, and generate the sensitivity level corresponding to each field. The result determination unit is used to determine the sensitivity level that meets the highest preset level condition among each field as the privacy classification result of the creation event fragment; Accordingly, the commitment generation module 12 may specifically include: The summary generation unit is used to generate a displayable summary for the fields whose sensitivity level is the public level; The input generation unit is used to generate a corresponding partial content summary and a proof input for generating a privacy protection process proof for the field whose sensitivity level is the verification level, based on a preset commitment generation rule; The field backup unit is used to perform encrypted backups of fields with the sensitivity level of the confidentiality level only locally.

[0048] In some specific implementations, the input generation unit may specifically include: The first summary generation subunit is used to perform summary calculation operations on the pre-version summary, post-version summary, and local difference summary of the fields at the verification level, and generate a summary with invisible content; the pre-version summary is the summary of the version before fragment processing, and the post-version summary is the summary of the version after fragment processing; The second summary generation subunit is used to perform summary calculation operations on the event type distribution, manual input density, version saving times, material import times, and AI-assisted call types of the fields at the verification level, and generate behavioral statistical summaries; The third summary generation subunit is used to determine the corresponding local content summary based on the content-invisible summary and the behavior statistics summary; The input generation subunit is used to combine the privacy classification result, quality marker, and AI-assisted call boundary summary of the fields of the verification level to generate the proof input.

[0049] In some specific implementations, the credibility generation module 14 may specifically include: The first penalty unit is used to determine a segment gap penalty item based on the portion of the time interval between adjacent creative event segments that exceeds the preset maximum normal segment interval; wherein, the segment gap penalty item is used to reduce the credibility of the creative event segment with abnormal time gaps; The second penalty unit is used to determine an automatic input anomaly penalty item based on the similarity between the input rhythm vector and the preset automatic input rhythm vector, and the coefficient of variation of the input rhythm vector; wherein, the automatic input anomaly penalty item is used to reduce the credibility of the creative continuity when the similarity is higher than the similarity threshold and the coefficient of variation is lower than the variation threshold. The first credibility determination unit is used to determine the credibility of the creative continuity based on the artificial event density of each creative event segment, the time interval between adjacent creative event segments, the segment gap penalty item, and the automatic input anomaly penalty item, using a first preset credibility calculation formula. The first preset credibility calculation formula is: ; in, This indicates the credibility of the stated creative continuity. This represents the Sigmoid normalization function, where n represents the number of creation event fragments. This represents the effective artificial event density of the i-th creative event fragment. This indicates the time interval between adjacent creative event segments. Indicates the average of time intervals. This indicates a fragment gap penalty term. This indicates that the penalty for errors will be automatically entered.

[0050] In some specific implementations, the credibility generation module 14 may specifically include: The second credibility determination unit is used to determine the credibility of the AI-assisted call boundary by using a second preset credibility calculation formula, based on the difference between the amount of manual modification, the amount of AI output, the distribution of AI output summary and the distribution of manually modified summary in the AI-assisted call boundary summary. The second preset credibility calculation formula is: ; in, This indicates the credibility of the AI-assisted invocation boundary. This represents the amount of effective modifications made by humans after the AI ​​output. Indicates the amount of AI output. This indicates a smoothing term to prevent the denominator from being zero. Indicates the Kullback-Leibler divergence. This indicates the distribution of AI output summaries. This indicates the distribution of the abstracts after manual modification. This indicates the proportion of large, unexplained output segments.

[0051] In some specific implementations, the proof package generation module 15 may specifically include: The category and granularity determination unit is used to determine the data categories that can be disclosed and the corresponding data disclosure granularity based on the target rights confirmation scenario; wherein, the target rights confirmation scenario includes any one or more of the following: rights confirmation and registration scenario, platform review scenario, ownership dispute scenario, and infringement evidence scenario; The proof extraction unit is used to extract corresponding privacy-preserving process proofs from the local summary commitment, the creation continuity chain, the AI-assisted call boundary summary, and the time anchor based on the data category and the data disclosure granularity; wherein, the privacy-preserving process proof is used to demonstrate that each chain node is generated and is valid by the corresponding creation event fragment without disclosing the detailed data of the creation event fragment; A credibility judgment unit is used to determine whether the credibility of the creation continuity is greater than or equal to a first preset credibility threshold, and whether the credibility of the AI-assisted call boundary is greater than or equal to a second preset credibility threshold, so as to obtain the corresponding credibility judgment result. The proof package generation unit is used to generate a creation process proof package corresponding to the target work based on the extracted privacy protection process proof and the credibility determination result.

[0052] Furthermore, embodiments of this application also disclose an electronic device, Figure 5 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the privacy-protected rights verification method of the creation process disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0053] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0054] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk, or optical disk, etc. The resources stored thereon can include an operating system 221, computer programs 222, etc., and the storage method can be temporary storage or permanent storage.

[0055] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the privacy-protected rights verification method for the creation process executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0056] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned privacy-preserving method for proving ownership of a creation process. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0057] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0058] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0059] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0060] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0061] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A privacy-protected method for proving ownership during the creative process, characterized in that, include: The creation process events generated during the creation of the target work are obtained, and the creation process events are divided into various creation event segments based on preset segmentation conditions. The creation event segments are then classified for privacy to generate corresponding privacy classification results. A partial summary commitment is generated based on the privacy classification result for the creation event fragment; the partial summary commitment is used to characterize the version evolution relationship of the creation event fragment without disclosing the detailed data of the creation event fragment. Each of the creation event fragments is transformed into chain nodes, and based on version evolution relationships and / or time order, each of the chain nodes is connected into a creation continuity chain; wherein, the chain nodes in the creation continuity chain are bound to the local summary commitment, time anchor point and AI-assisted call boundary summary of the corresponding creation event fragment, and the AI-assisted call boundary summary is a privacy representation of the relationship between AI-assisted call and manual modification in the corresponding creation event fragment; The credibility of the creative continuity of the target work is determined based on the creative continuity chain, and the credibility of the corresponding AI-assisted invocation boundary is determined based on the AI-assisted invocation boundary summary. Based on the preset privacy proof strategy corresponding to the target rights confirmation scenario, a privacy protection process proof is generated based on the creation continuity chain, the creation continuity credibility, the AI-assisted call boundary credibility, and the preset credibility threshold, and a creation process proof package containing the privacy protection process proof is output.

2. The privacy-protected rights confirmation method for the creation process according to claim 1, characterized in that, The process of acquiring creative process events generated during the creation of the target work is divided into various creative event segments based on preset segmentation conditions, including: During the creation process of the target work, events of the creation process are collected in real time and written to the local event cache; In the local event cache, if the collected creation process events meet any of the triggering conditions in the preset segmentation conditions, then the continuous creation process events that meet the triggering conditions are extracted to construct the corresponding creation event segments; wherein, the triggering conditions include any one or more of the following: reaching a preset time window, reaching a preset number of key operations, reaching a version node, and a work area switch.

3. The privacy-protected rights confirmation method for the creation process according to claim 1, characterized in that, The privacy classification results include public, verified, and confidential levels. The process of performing privacy classification on the creation event fragments to generate corresponding privacy classification results includes: The sensitivity level of each field contained in the creative event fragment is determined according to the preset sensitivity level classification conditions, and the sensitivity level corresponding to each field is generated. The sensitivity level that meets the highest preset level condition among all fields is determined as the privacy classification result of the creation event fragment; Accordingly, the commitment to generate a partial summary of the creation event fragment based on the privacy classification result includes: Generate a displayable summary for the fields whose sensitivity level is public. Based on preset commitment generation rules, for fields with the sensitivity level of the verification level, a corresponding partial content summary and proof input for generating proof of the privacy protection process are generated; Fields with the sensitivity level set to the confidentiality level are only encrypted and backed up locally.

4. The privacy-protected rights confirmation method for the creation process according to claim 3, characterized in that, The process of generating a local content digest and proof input for generating a privacy protection process proof for fields with a sensitivity level of the verification level based on preset commitment generation rules includes: The pre-version summary, post-version summary, and local difference summary of the fields at the verification level are processed to generate a summary with invisible content; the pre-version summary is the summary of the version before fragment processing, and the post-version summary is the summary of the version after fragment processing. The event type distribution, human input density, version saving times, material import times, and AI-assisted call types of the fields at the verification level are summarized to generate a behavioral statistical summary. Based on the content-invisible summary and the behavior statistics summary, a corresponding local content summary is determined; The proof input is generated by combining the privacy classification result, quality marker, and AI-assisted call boundary summary of the fields of the verification level.

5. The privacy-protected rights confirmation method for the creation process according to claim 1, characterized in that, Determining the credibility of the creative continuity of the target work based on the creative continuity chain includes: A segment gap penalty term is determined based on the portion of the time interval between adjacent creative event segments that exceeds the preset maximum normal segment interval; wherein, the segment gap penalty term is used to reduce the credibility of creative event segments with abnormal time gaps; An automatic input anomaly penalty term is determined based on the similarity between the input rhythm vector and the preset automatic input rhythm vector, and the coefficient of variation of the input rhythm vector; wherein, the automatic input anomaly penalty term is used to reduce the credibility of the creative continuity when the similarity is higher than a similarity threshold and the coefficient of variation is lower than a variation threshold. The credibility of the creative continuity is determined by the first preset credibility calculation formula, based on the artificial event density of each creative event segment, the time interval between adjacent creative event segments, the segment gap penalty item, and the automatic input anomaly penalty item. The first preset credibility calculation formula is: ； in, This indicates the credibility of the stated creative continuity. This represents the Sigmoid normalization function, where n represents the number of creation event fragments. This represents the effective artificial event density of the i-th creative event fragment. This indicates the time interval between adjacent creative event segments. Indicates the average of time intervals. This indicates a fragment gap penalty term. This indicates that the penalty for errors will be automatically entered.

6. The privacy-protected method for proving ownership of a creation process according to claim 1, characterized in that, The step of determining the corresponding AI-assisted invocation boundary credibility based on the AI-assisted invocation boundary summary includes: The credibility of the AI-assisted call boundary is determined by the second preset credibility calculation formula based on the difference between the amount of manual modification, the amount of AI output, the distribution of AI output summary and the distribution of manually modified summary in the AI-assisted call boundary summary. The second preset credibility calculation formula is: ； in, This indicates the credibility of the AI-assisted invocation boundary. This represents the amount of effective modifications made by humans after the AI ​​output. Indicates the amount of AI output. This indicates a smoothing term to prevent the denominator from being zero. Indicates the Kullback-Leibler divergence. This indicates the distribution of AI output summaries. This indicates the distribution of the abstracts after manual modification. This indicates the proportion of large, unexplained output segments.

7. The privacy-protected rights confirmation method for the creation process according to claim 1, characterized in that, The method, based on the preset privacy proof strategy corresponding to the target rights confirmation scenario, generates a privacy protection process proof based on the creation continuity chain, the creation continuity credibility, the AI-assisted call boundary credibility, and a preset credibility threshold, and outputs a creation process proof package containing the privacy protection process proof, including: The categories of data that can be disclosed and the corresponding granularity of data disclosure are determined based on the target rights confirmation scenarios; wherein, the target rights confirmation scenarios include any one or more of the following: rights confirmation and registration scenarios, platform review scenarios, ownership dispute scenarios, and infringement evidence scenarios; Based on the data category and the data disclosure granularity, corresponding privacy protection process proofs are extracted from the local summary commitment, the creation continuity chain, the AI-assisted call boundary summary, and the time anchor; wherein, the privacy protection process proof is used to demonstrate that each chain node is generated and valid by the corresponding creation event fragment without disclosing the detailed data of the creation event fragment; Determine whether the credibility of the creation continuity is greater than or equal to a first preset credibility threshold, and whether the credibility of the AI-assisted call boundary is greater than or equal to a second preset credibility threshold, so as to obtain the corresponding credibility determination result; Based on the extracted privacy protection process proof and the credibility determination result, a corresponding creation process proof package for the target work is generated.

8. A privacy-protecting rights confirmation device for the creative process, characterized in that, include: The result generation module is used to acquire the creation process events generated during the creation of the target work, divide the creation process events into various creation event segments based on preset segment division conditions, and perform privacy classification on the creation event segments to generate corresponding privacy classification results. The commitment generation module is used to generate a partial summary commitment of the creation event fragment based on the privacy classification result; the partial summary commitment is used to characterize the version evolution relationship of the creation event fragment without disclosing the detailed data of the creation event fragment. The node connection module is used to convert each of the creation event fragments into chain nodes, and connect each of the chain nodes into a creation continuity chain based on version evolution relationship and / or time order; wherein, the chain nodes in the creation continuity chain are bound to the local summary commitment, time anchor point and AI-assisted call boundary summary of the corresponding creation event fragment, and the AI-assisted call boundary summary is a privacy representation of the relationship between AI-assisted call and manual modification in the corresponding creation event fragment; The credibility generation module is used to determine the credibility of the creation continuity of the target work based on the creation continuity chain, and to determine the corresponding credibility of the AI-assisted call boundary based on the AI-assisted call boundary summary. The proof package generation module is used to generate a privacy protection process proof based on the preset privacy proof strategy corresponding to the target rights confirmation scenario, the creation continuity chain, the creation continuity credibility, the AI-assisted call boundary credibility, and the preset credibility threshold, and output a creation process proof package containing the privacy protection process proof.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the privacy-preserving method for proving ownership of the creation process as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store computer programs; wherein, when the computer programs are executed by a processor, they implement the privacy-protected method for proving ownership of the creation process as described in any one of claims 1 to 7.

Images