Supply chain analysis for artificial intelligence (AI) generated source code

Abstract

Source code generated by an Artificial Intelligence (AI) algorithm is analyzed to identify a snippet of source code generated by the AI algorithm that is the same or similar to source code used to train the AI algorithm. For example, a specific function in a component in the source code used to train the AI algorithm is identified in the snippet of source code generated by the AI algorithm. A determination is made to see if the identified snippet of source code generated by the AI algorithm is used in a software application. A bill-of-materials for the software application is generated that comprises information associated with the identified snippet of source code. In other words, the bill-of-materials for the software application includes a bill-of-materials for the AI generated source code in the software application.

Description

Supply Chain Analysis for Artificial Intelligence (Al) Generated Source CodeFIELD

[0001] The disclosure relates generally to managing software bill-of-materials and particularly to management of software bill-of-materials for source code generated by Al algorithms.BACKGROUND

[0002] One of the key issues today is to be able to track the supply chain of a product to produce a complete software bill-of-materials in order to identify issues like potential security risks within the software. For example, the SolarWinds hack in 2020 that caused a breach of more than 30,000 private and public organizations was due to a supply chain vulnerability. Nowadays, having a validated software bill-of-materials for a software product is critical. For example, the United States government is actively pushing for stronger software bill-of-materials (e.g., see https: / / www.cisa.gov / sbom). What is needed is a way to generate an accurate bill-of-materials for all software, which cannot be done today with the current software tracking tools.SUMMARY

[0003] These and other needs are addressed by the various embodiments and configurations of the present disclosure. The present disclosure can provide a number of advantages depending on the particular configuration. These and other advantages will be apparent from the disclosure contained herein.

[0004] Source code generated by an Artificial Intelligence (Al) algorithm is analyzed to identify a snippet of source code generated by the Al algorithm that is the same or similar to source code used to train the Al algorithm. For example, a specific function in a component in the source code used to train the Al algorithm is identified in the snippet of source code generated by the Al algorithm. A determination is made to see if the identified snippet of source code generated by the Al algorithm is used in a software application. A bill-of-materials for the software application is generated that comprises information associated with the identified snippet of source code. In other words, the bill- of-materials for the software application includes a bill-of-materials for the Al generatedsource code in the software application that associates additional metadata for tracking provenance of the Al generated code.

[0005] The phrases "at least one", "one or more", “or”, and "and / or" are open-ended expressions that are both conjunctive and disjunctive in operation. For example, each of the expressions "at least one of A, B and C", "at least one of A, B, or C", "one or more of A, B, and C", "one or more of A, B, or C", "A, B, and / or C", and "A, B, or C" means A alone, B alone, C alone, A and B together, A and C together, B and C together, or A, B and C together.

[0006] The term "a" or "an" entity refers to one or more of that entity. As such, the terms "a" (or "an"), "one or more" and "at least one" can be used interchangeably herein. It is also to be noted that the terms “comprising”, “including”, and “having” can be used interchangeably.

[0007] The term “automatic” and variations thereof, as used herein, refers to any process or operation, which is typically continuous or semi-continuous, done without material human input when the process or operation is performed. However, a process or operation can be automatic, even though performance of the process or operation uses material or immaterial human input, if the input is received before performance of the process or operation. Human input is deemed to be material if such input influences how the process or operation will be performed. Human input that consents to the performance of the process or operation is not deemed to be “material”.

[0008] Aspects of the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a “circuit,” “module” or “system.” Any combination of one or more computer readable medium(s) may be utilized. The computer readable medium may be a computer readable signal medium or a computer readable storage medium.

[0009] A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, aportable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.

[0010] A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0011] The terms “determine”, “calculate” and “compute,” and variations thereof, as used herein, are used interchangeably, and include any type of methodology, process, mathematical operation, or technique.

[0012] The term “means” as used herein shall be given its broadest possible interpretation in accordance with 35 U.S.C., Section 112(f) and / or Section 112, Paragraph 6. Accordingly, a claim incorporating the term “means” shall cover all structures, materials, or acts set forth herein, and all of the equivalents thereof. Further, the structures, materials or acts and the equivalents thereof shall include all those described in the summary', brief description of the drawings, detailed description, abstract, and claims themselves.

[0013] As described herein an in the claims, a “bill-of-materials” may relate to software, firmware, shell scripts, executable code (e.g., a binary), a combination of these, and / or the like. A bill-of-materials may be a record for a single component, for multiple components, for a piece of source code, and / or the like. In other words, any software / firmware associated with a software application.

[0014] The preceding is a simplified summary to provide an understanding of some aspects of the disclosure. This summary is neither an extensive nor exhaustive overview of the disclosure and its various embodiments. It is intended neither to identify key or critical elements of the disclosure nor to delineate the scope of the disclosure but topresent selected concepts of the disclosure in a simplified form as an introduction to the more detailed description presented below. As will be appreciated, other embodiments of the disclosure are possible utilizing, alone or in combination, one or more of the features set forth above or described in detail below. Also, while the disclosure is presented in terms of exemplary embodiments, it should be appreciated that individual aspects of the disclosure can be separately claimed.BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Fig. l is a block diagram of a first illustrative system for creating a bill-of- materials for Artificial Intelligence (Al) generated source code.

[0016] Fig. 2 is a block diagram of a second illustrative system for creating a bill-of- materials for Al generated source code.

[0017] Fig. 3 is a flow diagram of a process creating a bill-of-materials for Al generated source code.

[0018] Fig. 4 is a flow diagram of a process for identifying source(s) of Al generated source code using vectors.

[0019] Fig. 5 is a flow diagram of a process for identifying source(s) of Al generated source code using hashes of snippets of source code.

[0020] Fig. 6 is a flow diagram of a process for identifying source(s) of Al generated source code using snippets of source code.

[0021] Fig. 7 is a flow diagram of a process for determining licenses associated with Al generated source code.

[0022] Fig. 8 is a flow diagram of a process for determining information associated with Al generated source code.

[0023] Fig. 9 is a flow diagram of a process for determining information associated with training an Al algorithm.

[0024] Fig. 10 is a diagram of a blockchain for storing information associated with training an Al algorithm.

[0025] Fig. 11 is a diagram of a blockchain for storing information associated with Al generated source code.

[0026] In the appended figures, similar components and / or features may have the same reference label. Further, various components of the same type may be distinguished by following the reference label by a letter that distinguishes among the similar components. If only the first reference label is used in the specification, the description is applicable toany one of the similar components having the same first reference label irrespective of the second reference label.DETAILED DESCRIPTION

[0027] Fig. 1 is a block diagram of a first illustrative system 100 for creating a bill-of- materials 127 for Artificial Intelligence (Al) generated source code 123. The first illustrative system 100 comprises communication devices 101A-101N, a network 110, a server 120, and a distributed ledger 130. In addition, users 102A-102N are shown for convenience.

[0028] The communication devices 101 A- 10 IN can be or may include any user device that can communicate on the network 110, such as a Personal Computer (PC), a cellular telephone, a Personal Digital Assistant (PDA), a tablet device, a notebook device, a laptop computer, a smartphone, and / or the like. As shown in Fig. 1, any number of communication devices 101 A- 10 IN may be connected to the network 110, including only a single communication device 101. The users 102A-102N use the communication devices 101A-101N to communicate to the server 120.

[0029] The network 110 can be or may include any collection of communication equipment that can send and receive electronic communications, such as the Internet, a Wide Area Network (WAN), a Local Area Network (LAN), a packet switched network, a circuit switched network, a cellular network, a combination of these, and the like. The network 110 can use a variety of electronic protocols, such as Ethernet, Internet Protocol (IP), Hyper Text Transfer Protocol (HTTP), Web Real-Time Protocol (Web RTC), and / or the like. Thus, the network 110 is an electronic communication network configured to carry messages via packets and / or circuit switched communications.

[0030] The server 120 may be any type of device that can be accessed by the communication devices 101 A- 10 IN. The server 120 may be used by the user 102 to develop and generate different types of source code (e.g., the Al generated source code 123). The server 120 comprises training source code 121, an Al algorithm 122, the Al generated source code 123, a similarity algorithm 124, a blockchain / database manager 125, a database 126, a bill-of-materials for other software 1270, a software application 128. In one embodiment, the server 120 may be a communication device 101 (e.g., a user’s laptop computer). In addition, the different elements 121-128 may be distributed between a communication device 101 and the server 120.

[0031] The training source code 121 is any source code used to train the Al algorithm 122. The training source code 121 is shown on the server 120. However, the training source code 121 may reside in other places, such as an open-source repository (e.g., GitHub), a development server, and / or the like. The training source code 121 may be proprietary source code, open-source code, third party source code, public domain source code, and / or the like. The training source code 121 may be subject to various types of licenses, such as proprietary licenses, open-source licenses (e.g., the GPL license, the MIT license, the BSD license, the Apache license, etc.), public domain licenses, copyright licenses, and / or the like. The training source code 121 may be in various programming languages, such as Java, C, C++, assembly, Hyper Text Markup Language (HTML), Cobal, Python, JavaScript, and / or the like.

[0032] The Al algorithm 122 may be any type of Al algorithm 122 that can generate source code, such as OpenAICodex, Tabnine, Code5T, Polycoder, GetHub Copilot, and / or the like. The Al algorithm 122 is trained using the training source code 121. The Al algorithm 122 generates the Al generated source code 123 based on the training source code 121.

[0033] The similarity algorithm 124 may comprise one or more algorithms that are used to identify similarities between the training source code 121 and the Al generated source code 123. The similarity algorithm 124 may comprises various types of algorithms that use vectors, hashes, snippets, and / or the like. The similarity algorithm 124 may run multiple algorithms in parallel. For example, the similarity algorithm may comprise a vector algorithm and a hashing algorithm that run in parallel.

[0034] The similarity algorithm 124 is used to overcome problems with existing Al algorithm 122 that generate source code. With the complexity of Large Language Models (LLM) Al algorithms 122, it is currently not possible to identify what training source code 121 produced the actual Al generated source code 123. This is because of the many layers within the LLMs and the fact that each layer does not retain the information of the previous layer.

[0035] The blockchain / database manager 125 may be any hardware coupled with software that can manage the Al processes described herein and manage the retrieval / storage of information in the database 126 and / or the distributed ledger 130 / blockchain 131. The blockchain / database manager 125 can be used create the blockchains 131 in the distributed ledger 130. In addition, the blockchain / database manager 125 may add information to the bill-of-materials 127BC in the blockchains 131,add blocks to the blockchains 131 in the distributed ledger 130, and / or the like. Likewise, the blockchain / database manager 125 may add information associated with the bill-of- materials 127DB in the database 126.

[0036] The database 126 may be any type of database 126 that can store information, such as a relational database, a hierarchical database, an analytical database, a file system, and / or the like. The database 126 further comprises the bill-of-materials 127DB. The bill- of-materials 127DB may be stored in different structures / tables in the database 126.

[0037] The bill-of-materials for other software 1270 is a bill-of-materials of source code that is not generated by the Al algorithm 122. For example, the bill-of-materials for other software 1270 may be based on source code / components that are developed by the users 102A-102N, from a third-party, from open-source code that is not used to train the Al algorithm 122, from a proprietary code base, and / or the like. The bill-of-materials for the other software 1270 may comprise multiple bill-of-materials for each component.

[0038] The software application 128 may comprise any type of software application 128, such as a web application, a security application, an embedded application, a financial application, a database application, a communication application, an email application, a networked application, and / or the like. The software application 128 may comprise sub-portions of an application such as a component, library, framework, API, or the like. The software application 128 may comprise a firmware, a shell scripts, software, instruction code, executable code, a combination of these, and / or the like.

[0039] The distributed ledger 130 comprises two or more nodes that store copies of the blockchains 131. The nodes use a consensus vote for adding blocks to the blockchains 131. The distributed ledger 130 is shown separate from the server 120. However, the server 120 may be a node that has a copy of the blockchain 131 in the distributed ledger 130. The blockchains 131 comprise the bill-of-materials 127BC that is made up of different blocks that are stored as part of the blockchains 131 in the distributed ledger 130.

[0040] The distributed ledger 130 may be a private distributed ledger 130 or a public distributed ledger 130. For example, the distributed ledger 130 may be a private distributed ledger 130 that is owned by a single entity (e.g., a corporation). Alternatively, the distributed ledger 130 may be a public distributed ledger 130 that can be publicly viewed by anyone. The distributed ledger 130 may be a semi-private distributed ledger 130.

[0041] Fig. 2 is a block diagram of a second illustrative system 200 for creating a bill- of-materials 127 for Al generated source code 123. When the Al algorithm 122 is used togenerate the Al generated source code 123, various aspects of the Al algorithm 122 / Al generated source code 123 are tracked and stored in a distributed ledger 130 / blockchains 131A-131N and / or the database 126.

[0042] Fig. 2 comprises the training source code 121, the Al algorithm 122, the Al generated source code 123, the similarity algorithm 124, the bill-of-materials for other software 1270, the blockchain / database manager 125, the distributed ledger 130 that includes the blockchains 131A-131N (that have a replicated bill-of-materials 127BCA- 127BCN), and the database 126 that has the bill-of-materials 127DB. In Fig. 2, the training source code 121 is used to train the Al algorithm 122, which in turn generates the Al generated source code 123 based on input parameters from the user 102 (or could be from an automated source or another Al algorithm). The similarity algorithm 124 determines similarities between the Al generated source code 123 and the training source code 121. The blockchain / database manager 125 then takes different types of information from the training source code 121, the Al algorithm, 122 the generated source code 123, the similarity algorithm 124, the bill-of-materials for other software 1270, and the user input parameters to create the bill-of-materials 127BCA-127BCN and / or bill-of-materials 127DB. While described using a distributed ledger 130, the bill-of-materials 107BC may reside in a single blockchain 127BC.

[0043] Fig. 3 is a flow diagram of a process creating a bill-of-materials 127BC / 127DB for Al generated source code 123. Illustratively, the communication devices 101A-101N, the server 120, the training source code 121, the Al algorithm 122, the Al generated source code 123, the similarity algorithm 124, the blockchain / database manager 125, the database 126, the bill-of-materials 127DB, the a bill-of-materials for other software 1270, the distributed ledger 130, the blockchains 131A-131N, and the bill-of-materials 127BCA- 127BCN are stored-program-controlled entities, such as a computer or microprocessor, which performs the method of Figs. 3-11 and the processes described herein by executing program instructions stored in a computer readable storage medium, such as a memory (i.e., a computer memory, a hard disk, and / or the like). Although the methods described in Figs. 3-11 are shown in a specific order, one of skill in the art would recognize that the steps in Figs. 3-11 may be implemented in different orders and / or be implemented in a multi -threaded environment. Moreover, various steps may be omitted or added based on implementation.

[0044] The process starts in step 300. The similarity algorithm 124 gets, in step 302, the Al generated source code 123. The similarity algorithm 124 analyzes the Al generatedsource code 123 to identify one or more snippets of the Al generated source code 123 that are the same or similar to the training source code 121 in step 304. For example, the similarity algorithm 124 may be a vector-based algorithm that takes snippets of the training source code 121 and Al generated source code 123 to produce vectors (e.g., floating-point vectors). The vectors from the training source code 121 are compared to the vectors of the Al generated source code 123 to identify exact matches and / or similarities (e.g., identical floating-point vectors or close floating-point vectors).

[0045] The similarity algorithm 124 determines, in step 306, if there are any matching / similar snippets in the Al generated source code 123. If there were not any matching / similar snippets in step 306, the process goes to step 316. Otherwise, if there are one or more matching / similar snippets in step 306, the blockchain / database manager 125 determines, in step 308, if the one or more identified snippets are in a software application 128 that requires a bill-of-materials 127BC / 127DB.

[0046] If there are no matching snippets in the software application 128 in step 308, the process goes to step 316. Otherwise, if there are one or more matching / similar snippets in step 308, the blockchain / database manager 125 determines, in step 310, if there are other software / firmware components in the software application 128. If there are not any other software / firmware components in the software application 128 or there is not a bill-of- materials for the other software 1270 in step 310, the process goes to step 314. If there are other software / firmware components and there is a bill-of-materials 1270 for the other software / firmware component(s) in step 310, the blockchain / database manager 125 gets the bill-of-materials for the other softwarel27O in step 312.

[0047] The blockchain / database manager 125 generates the bill-of-materials 127BC / 127DB for the software application 128 in step 314 to include the evidence of provenance of the application and for the Al generated source code 123. The bill-of- materials 127BC / 127DB may include the bill-of-materials for other software 1270, information associated with a specific version of the Al algorithm 122, information associated with a specification version of the Al generated source code 123, information associated with a specific time that input data was used to generated the Al generated source code 123 (e.g., when nothing else has changed (i.e., the training source code 121, the Al algorithm version, the input data / parameters used to generate the Al generated source code 123, etc. are the same)), information associated with the input data / parameters used to generate the Al generated source code 123, a specific version of the Al algorithm 122, a specific version of the training source code 121, information about the Al algorithm122, license information about the Al generated source code 123, a likely license associated with the identified snippet of source code generated by the Al algorithm 122, trust information associated with the identified snippet of source code generated by the Al algorithm 122, a hash of the snippet of Al generated source code 123, the snippet of source code generated by the Al algorithm 122, vector(s) generated by the similarity algorithm 124, the Al generated source code 123, information associated with the software application 128, testing information, source code removal information, release information, and / or the like. This may include any links to this information. The process then goes to step 316.

[0048] Because all this information is tracked, the bill-of-materials 127BC / 127DB for the software application 128 can identify critical information. For example, the blockchain / database manager 124 may identify the Al generated source code 123, the version of the Al algorithm 122, the training source code 121 used to train the Al algorithm 122, the input param eters / data, the time the Al generated source code 123 was created (because different Al generated source code 123 may be generated when nothing else has changed), the user who generated the input, and / or the like to properly create the bill-of-materials 127BC / 127DB.

[0049] The bill-of-materials 127BC / 127DB generated in step 314 may comprise information for Al generated source code 123 from different versions of Al generated source code 123 where only the input data / parameters have changed. For example, component A from input parameters A and component B from input parameters B may be used in the same software application 128 / bill-of-materials 127BC / 127DB. Another example is where the bill-of-materials 127BC / 127DB comprises two or more sets of Al generated source code 123 that came from two or more sets of training source code 121. The bill-of-materials 127BC / 127DB may come from two or more sets of Al generated source code 123 where the only difference is a different time (nothing else changed in the development environment) that the Al generated source code 123 was generated. The bill- of-materials 127BC / 127DB may include where the Al generated source code 123 has been modified. For example, the user 102 may modify the Al generated source code 123 by adding source code to the Al generated source code 123, removing source code from the Al generated source code 123, and / or the like.

[0050] The blockchain / database manager 125 determines, in step 316, if the process is complete. If the process is not complete in step 316, the process goes back to step 302. Otherwise, if the process is complete in step 316, the process ends in step 318.

[0051] Fig. 4 is a flow diagram of a process for identifying source(s) of Al generated source code 123 using vectors. The process of Fig. 4 is an exemplary embodiment of step 304 of Fig. 3 where the similarity algorithm 124 is a vector-based similarity algorithm 124.

[0052] After getting the Al generated source code 123 in step 302, the similarity algorithm 124 generates vectors for the Al generated source code 123 (e.g., by creating floating-point vectors for snippets of the Al generated source code 123) in step 400. The similarity algorithm 124 generates vectors for the training source code 121 (e.g., by creating floating point vectors for snippets of the training source code 121) in step 402. The similarity algorithm 124 determines, in step 404, if there are any matched vector(s) / similar vectors in step 404. The determination of step 404 may use a threshold to determine similar vectors.

[0053] If there are no matched or similar vectors in step 404, the process goes to step 306. Otherwise, if there are one or more matched or similar vectors in step 404, the snippet(s) of the training source code 121 are identified based on the matched or similar vectors in step 406. The vector information / matching information is then stored off in step 408 and the process goes to step 306.

[0054] The stored off information of step 408 may include the matched vector(s) (e.g., the floating-point data), the matched snippets, the identified component s) in the training source code 121, the identified component s) in the Al generated source code 123, matching vector information generated by the similarity algorithm 124, license information, links to any of the above, and / or the like. The information stored off in step 408 is used to generate the bill-of-materials 127BC / 127DB in step 314.

[0055] Fig. 5 is a flow diagram of a process for identifying source(s) of Al generated source code 123 using hashes of snippets of source code. The process of Fig. 5 is an exemplary embodiment of step 304 of Fig. 3 where the similarity algorithm 124 is a snippet hash-based similarity algorithm 124.

[0056] After getting the Al generated source code 123 in step 302, the similarity algorithm 124 breaks the Al generated source code 123 into snippets and then generates hashes of the snippets in step 500. The similarity algorithm 124 breaks the training source code 121 into snippets and then generates hashes of the snippets in step 502.

[0057] The similarity algorithm 124 identifies, in step 504, if there are any hash matches. For example, a match is where a hash of a snippet of the training source code 121 is the same as a hash for a snippet the Al generated source code 123. If there are notany matches in step 504, the process goes to step 306. Otherwise, if there is one or more matches in step 504, the similarity algorithm 124 identifies the snippet(s) in the training source code 121 that match the snippets from the Al generated source code 123 in step 506. The matching snippets, hash information, license information, and / or other matching information are then stored off in step 508. The information stored off in step 508 is used to generate the bill-of-materials 127BC / 127DB in step 314.

[0058] Fig. 6 is a flow diagram of a process for identifying source(s) of Al generated source code 123 using snippets of source code. The process of Fig. 6 is an exemplary embodiment of step 304 of Fig. 3 where the similarity algorithm 124 is a snippet-based similarity algorithm 124.

[0059] After getting the Al generated source code 123 in step 302, the similarity algorithm 124 breaks the Al generated source code 123 into snippets in step 600. The similarity algorithm 124 breaks the training source code 121 into snippets in step 602.

[0060] The similarity algorithm 124 identifies, in step 604 if there are any snippet matches. For example, a match is where a snippet of the training source code 121 is the same or similar to a snippet from the Al generated source code 123. If there is not a match in step 604, the process goes to step 306. Otherwise, if there is one or more matches in step 604, the similarity algorithm 124 identifies the snippet(s) in the training source code 121 that match the snippets from the Al generated source code 123 in step 606. The matching snippets, licensing information, and / or other matching information are then stored off in step 608. The information stored off in step 608 is used to generate the bill- of-materials 127BC / 127DB in step 314.

[0061] The processes of Figs. 4-6 may be fine-tuned by changing the window size (i.e., the size of the snippets (e.g., lines / characters)). In addition, the processes of Figs. 4-6 may use over-lapping window / snippets. For example, for two consecutive snippets (a first and second snippet), there may be one overlapping snippet that has a portion of the first snippet and a portion of the second snippet. In addition, the threshold may be lower based on the license type. For example, if the license is GPL, a lower threshold may be used versus a non-viral license (e.g., a MIT license).

[0062] In addition, the system could account for snippets of code that are considered trivial code (i.e., common place snippets), which are considered non-copyrightable code based on identification of the snippets. For example, the similarity algorithm 124 could prefer snippets that are less frequent (e.g., only one instance) in the training set, over morecommon snippets. This could be achieved through an "inverse document frequency" mechanism (e.g., used today in the TF-IDF search algorithm).

[0063] Fig. 7 is a flow diagram of a process for determining licenses associated with Al generated source code 123. The process of Fig. 7 goes between step 306 (yes branch) and step 308. After identifying one or more matching / similar code snippets in step 306, the blockchain / database manager 125 determines license(s) associated with each matched / similar snippet in the training source code 121 in step 700. There can be more than one license associated with a matched snippet. For example, an open-source component may be licensed under multiple licenses, such as the GPL V2 license and the Apache 2.0 license.

[0064] The blockchain / database manager 125 determines the amount of each of the snippets of the same type of license in comparison to the training source code 121 in step 702 (a likely license). For example, if there are two snippet matches that are for Apache 2.0, the matched amounts are added together (e.g., 2% of the training source code 121 for snippet A and 3% of the training source code 121 for snippet B would total 5%). Another example is where there is only a single snippet that is 2% of the training source code 121 and uses the MIT license, the Al generated source code 123 would be identified as being derived from 2% of the MIT licensed source code.

[0065] If there are multiple licenses associated with the same snippet, the likely licenses can be identified as being licensed for either. For example, if snippet one is identified for 2% of the training source code 121 and is licensed under the Apache 2.0 license and snippet two is licensed under the Apache 2.0 license and the GPL V2 license, the percentages may be shown as 2-4% Apache and 0-2% GPL V2 (a varying percentage). The user 102 may then be given the option to select which license to use so that the actual percentages for the likely licenses match the user selected licenses. While described using open-source code, this process could also apply to any type of license.

[0066] The likely license(s) are stored off in step 704. The stored off licenses may be stored off based on a threshold for generating the bill-of-materials 127BC / 127DB. For example, the threshold may be .5% of the training source code 121. The threshold may be user defined or may be preset.

[0067] The likely licenses may then be optionally displayed to the user 102 in step 706. In one embodiment, the likely licenses may be displayed based on the threshold in step 706. In addition, the threshold may be displayed. The user 102 may also have the optionto change the threshold to see what licenses would be covered under a specific threshold. The process then goes to step 308.

[0068] Fig. 8 is a flow diagram of a process for determining information associated with Al generated source code 123. The process starts in step 800. The Al algorithm 122 determines, in step 802, if a request to generate the Al generated source code 123 has been received. If a request has not been received to generate the Al generated source code 123, the process of step 802 repeats. Otherwise, if a request to generate the Al generated source code 123 in step 802, the blockchain / database manager 125 gets various types of information associated with the generation of the Al generated source code 123 in step 804. For example, the blockchain / database manager 125 may get information associated with the Al algorithm 122 (e.g., a version number), generated source code information (e.g., a version number), training source code information (e.g., versions / dates of components of the training source code 121), a time when the Al generated source code 123 was generated, input data / parameters (e.g., information input by the user 102 to generate the Al generated source code 123), license information associated with the training source code 121 (e.g., open-source license information, third-party license information, proprietary license information, and / or the like), user information, location information, and / or the like.

[0069] The information of step 804 is then stored off in step 806. The stored off information may be later used to create the bill-of-materials 127BC / 127DB as described in step 314.

[0070] The blockchain / database manager 125 determines, in step 808, if the process is complete. If the process is not complete in step 808, the process goes back to step 802. Otherwise, if the process is complete in step 808, the process ends in step 810.

[0071] Fig. 9 is a flow diagram of a process for determining information associated with training an Al algorithm 122. The process starts in step 900. The Al algorithm 122 determines, in step 902, whether to train the Al algorithm 122. If the Al algorithm 122 is not to be trained, the process of step 902 repeats.

[0072] Otherwise, if the Al algorithm 122 is to be trained in step 902, the blockchain / database manager 125 gets information associated with training the Al algorithm 122 in step 904. For example, the information may be information about the training source code 121 (e.g., version numbers, hashes of components, dates, etc.), license information about the training source code 121 (e.g., open-source license information, proprietary license information, third-party license information, and / or thelike), user information (e.g., the user 102 who initiated the training of the Al algorithm 122), the time the Al algorithm was trained, a country that training of the Al algorithm took place, and other information (e.g., the version of the Al algorithm 122 when trained), and / or the like.

[0073] The information of step 904 is then stored off in step 906. The stored off information may be later used to create the bill-of-materials 127BC / 127DB as described in step 314.

[0074] The blockchain / database manager 125 determines, in step 908, if the process is complete. If the process is not complete in step 908, the process goes back to step 902. Otherwise, if the process is complete in step 808, the process ends in step 910.

[0075] Fig. 10 is a diagram of a blockchain 131 for storing information associated with training an Al algorithm 122 for tracking of provenance. The blockchain 131 is an exemplary example of the replicated blockchains 131A-131N in the distributed ledger 130. The blockchain 131 comprises a genesis block 1000, an Al algorithm block 1001, a training code block 1002, a license block 1003, and a trust information block 1004. The blocks 1000-1004 are linked together by links 1010A-1010D. The links 1010A-1010D are links to hashes (not shown) that are traditionally used in blockchains 131.

[0076] The blockchain / database manager 125 creates the genesis block 1000. The genesis block 1000 is generated to start a new blockchain 131 for tracking the bill-of- materials 127BCA-127BCN for a software application 128. In Fig. 10, the bill-of- materials 127BCA-127BCN comprise the blocks 1000-1004. However, additional blocks may be added to the blockchain 131A-131N / bill-of-materials 127BCA-127BCN as described in Fig. 11.

[0077] Once the genesis block 1000 is created, the Al algorithm block 1001 is created in the blockchain 131. The Al algorithm block 1001 may include various types of information associated with the Al algorithm 122, such as the name of the Al algorithm 122 (e.g., Al CodeGenerator X), a version of the Al algorithm 122 (e.g., version 1.0), a creation date of the Al algorithm 122 (e.g., March 6, 2023), a hash of the Al algorithm 122, a user 102 executing the Al algorithm 122, source code used to create the Al algorithm 122, hashes of source code used to create the Al algorithm 122, a country of origin of the Al algorithm 122, and / or the like. In one embodiment, the genesis block 1000 may be the Al algorithm block 1001.

[0078] When the Al algorithm 122 is trained, the training code block 1002 may be added to the blockchain 131. The training code block 1002 can comprise various types ofinformation, such as the user 102 who trained the Al algorithm 122, the date the Al algorithm 122 was trained, links to the training source code 121 (e.g., to different class files), the actual training source code 121, hashes of the training source code 121, country of origin of the user 102 who trained the Al algorithm 122, a country the Al algorithm was trained in, and / or the like.

[0079] As part of the training process, licenses associated with the training source code 121 are identified. For example, an Apache 1.0 open-source license may be associated with a specific component of the training source code. The license block 1003 is added to the blockchain 131 based on the identified licenses. The license block 1003 contains all (or some) of the identified licenses associated with the training source code 121.

[0080] In addition, a trust information block 1004 can be added to the blockchain 131. The trust information block 1004 may include various trust aspects of trust of the training source code 121, such as supply malicious taint (e.g., a component of the training source code 121 came from a malicious supplier), supply counterfeit (e.g., a component is a counterfeit), supply hygiene risks (e.g., the quality of a development environment for the training source code 121), supplier financial stability risks (e.g., is a supplier of a component stable and likely to maintain the component), supplier organizational security risks (e.g., does the supplier of the component introduce security risks in a component), supplier susceptibility (e.g., does the supplier of the component have security issues), supplier quality culture risks (e.g., the supplier has a coder who constantly writes bad source code), supplier organization effectiveness risks (e.g., is the supplier inclined to introduce risks), supplier ethical risks (e.g., does the supplier use unethical practices), supplier external influences (e.g., is the supplier in a country that influences what is in a component), service quality risks, service resilience risks, service security risks, service integrity risks, and / or the like.

[0081] Although not shown, other blocks may be added to the blockchain 131. For example, a testing block may be added that has testing results of the training source code 121. The testing block may have information about tests of the Al generated source code 123, such as malware detected / fixed, bug fixed / not fixed, vulnerabilities removed, vulnerabilities not removed, and / or the like. The testing block may include any changes to the Al generated source code 123 that came as a result of the testing process. For example, the testing block may indicate that the file B in the Al generated source code 123 was changed because of an array overflow vulnerability.

[0082] Other types of blocks that may be added to the blockchain 131 may include a licenses filtered out block. The licenses filtered out block may be based on a filter that is used to filter the training source code 121.

[0083] Fig. 11 is a diagram of a blockchain 131 for storing information associated with Al generated source code 123 to track provenance. Fig. 11 is a continuation of the blockchain 131 described in Fig. 10. The blockchain 131 further comprises a user input block 1005, a generated source code block 1006, a likely license block 1007, a software application block 1008, and a release lock block 1009. The blocks 1005-1009 are linked together by links 1010E-1010N. The link 1011 links back to the genesis block 1000 of Fig. 10 to lock the blockchain 131. The link 1010E links back to the trust information block 1004 of Fig. 10.

[0084] Once the Al algorithm 122 is trained and the 102 user 102 now wants to generate the Al generated source code 123, additional information may be tracked in the blockchain 131 as shown in Fig. 11. As part of the process for generating the Al generated source code 123, the user 102 may provide various types of param eters / data to generate the Al generated source code 123. For example, the user 102 may provide input to create a shopping application that includes a products window, a shopping cart, a checkout window, a payment window, and an email process for sending emails to customers who purchase specific products. The input information may include snippets of source code. This results in the creation of the user input block 1005 (e.g. an Al input block). The user input block 1005 includes the user 102 (e.g., Jon Doe) who generated the Al generated source code 123, the date the Al generated source code 123 was generated (e.g., March 6, 2023), the input parameters (including snippets of source code), and / or the like.

[0085] The user input block 1005 may be where the input was not from a user 102, but from an Al or automated process. In this case, the block would be an Al input block.

[0086] The generated source code block 1006 is created when the Al generated source code 123 is generated. The generated source code block 1006 may comprise the actual Al generated source code 123, link(s) to the Al generated source code 123, hashes of the Al generated source code 123, a version number of the Al generated source code 123, a country of origin of the user 102 who initiated the generation of the Al generated source code 123, a country the Al generated source code 123 was generated in, and / or the like.

[0087] As the similarity algorithm 124 identifies the snippets in the training source code 121 as described in Fig. 3 / Fig. 7 and herein, the likely license block 1007 is added to the blockchain 131. The licenses in the likely license block 1007 are based on matching of thesnippets of the Al generated source code 123 to the training source code 121. In this example, there were three matching licenses: 1) Apache 8%, GPL V2 2%, and LGPL .01%. Although only open-source licenses are described, the likely license block 1007 may include other types of licenses. For example, a third-party license, a proprietary license, a copyright license, a public domain license, and / or the like may be identified. The likely licenses block 1007 may include a threshold that is used to filter the licenses. For example, if the filter was set at .02%, the LGPL license of .01% would not be included in the likely license block 1007, because the threshold of .02% is higher than the percentage of the LGPL license (.01%).

[0088] In addition, other blocks may be added that are not shown in Fig. 11. Another block that could be added is where the vectors (e.g., described in Fig. 4) for the similarity algorithm 124 are inserted as a block the blockchain 131. If hashes or snippets are used (e.g., as described in Figs 5-6), the identified hashes / snippets could be stored in the blockchain 131 in a corresponding hash / snippet block.

[0089] Once the Al generated source code 123 is placed into the software application 128, (e.g., a product for a release and / or during development), the software application block 1008 is created (or at a later time). The software application block 1008 may include the software application name (e.g., application Y), the software application version (e.g., version 3.6), the release date (September 3, 2023), links to the files in the software application 128, hashes of files in the software application 128, any changes made to the Al generated source code 123, information associated with any other source code / firmware associated with the software application 128, and / or the like. This may include if only a portion of the Al generated source code 123 is used. In this case, each file / snippet of the Al generated source code 123 that is used in the released version of the software application 128 is tracked in the software application block 1008.

[0090] In one embodiment, the release lock block 1009 may be added to the blockchain 131. The release lock block 1009 is a block that locks the blockchain 131 (i.e., it is the last block in the blockchain 131). The release lock block 1009 may point back to the genesis block 1000 with the link 1011. In one embodiment, the genesis block 1000 may have a link back to the release lock block 1009. If the release lock block 1009 is not used, the process may be repeated for different releases of the same software application 128 and / or for releases of other software applications 128 that use the Al generated source code 123 and / or new versions of the Al generated source code 123. Alternatively, each release of the software application 128 may be tracked in its own blockchain 131, in newbranches of the blockchain 131, and / or in new branches off the genesis block 1000 (e.g., a star blockchain 131). In one embodiment, instead of using the release lock block 1009, a release block may be used. In this example, the release block just has the release information and does not lock the blockchain 131. The release block does not have the link 1011 to the genesis block 1000.

[0091] Although not shown, the blockchain 131 may allow for removal of a file(s) (e.g., have a remove block). The removal block(s) may show the date, the user 102 who removed the file / component, version, and / or the like. Another type of block may be a modification block where the user 102 modified the Al generated source code 123 (e.g., a component in the Al generated source code 123). The modification block may identify the user 102 who modified the Al generated source code 123, a date the Al generated source code 123 was modified, changes to the Al generated source code 123, a location where the Al generated source code 123 was modified, and / or the like.

[0092] In addition, the information in different blocks 1000-1009 may be combined into a single block and / or into multiple blocks. For example, the software application block 1008 and the release lock block 1009 may be in a single block.

[0093] In addition, the order of how the different blocks 1001-1009 are added to the blockchain may vary. For example, the user input block 1006 and the generated source code block may be added in a reverse order.

[0094] While the above examples are discussed using a blockchain 131, the use of a blockchain 131 is not required. The information of the blocks 1000-1009 may be stored in different tables and / or different records in the database 126. For example, the information in each block 1000-1009 may be stored as a separate record in the database 126. In this example, the bill-of-materials 127DB is stored in the records of the database 126.

[0095] If the Al algorithm 122 is retrained using new training source code 121, the blockchain 131 may be updated with new training information (e.g., a new training code block 1002). The information could just include the delta. For example, if only a single training file / component was changed, added, or removed, the delta would be only in the new training code block 1002. In this example, there may be information indicating whether it was a change to a file, a new file, or a deleted file. Alternatively, when the Al algorithm 122 is retrained, all the new training information may be added as discussed above. In one embodiment, a new blockchain 131 or a branch off the genesis block 1000 may be created using the new training information.

[0096] In addition, the blockchain 131 may include blocks that have the other software that is included in the final software application 128 (e.g., the information from the bill-of- materials for other software 1270). This allows for a complete bill-of-materials 127BC that is stored in the blockchain 131.

[0097] The systems / processes described herein may be part of a Software as a Service (SaaS) platform where individual entities submit their data to create a centralized bill-of- materials 127. The centralized blockchain 131 could be a private blockchain 131, semiprivate blockchain 131, or a public blockchain 131.

[0098] In one embodiment, the training source code 121 and / or the modified Al generated source code 123 may be submitted via code tracking tools. For example, an Integrated Development Environment (IDE) may be used to manage the training source code 121 / AI generated source code 123. While this assumes a user 102 provides the input, other non-human processes may be used, such as automatic check in of the Al generated source code 123.

[0099] Examples of the processors as described herein may include, but are not limited to, at least one of Qualcomm® Snapdragon® 800 and 801, Qualcomm® Snapdragon® 610 and 615 with 4G LTE Integration and 64-bit computing, Apple® A7 processor with 64-bit architecture, Apple® M7 motion coprocessors, Samsung® Exynos® series, the Intel® Core™ family of processors, the Intel® Xeon® family of processors, the Intel® Atom™ family of processors, the Intel Itanium® family of processors, Intel® Core® i5- 4670K and i7-4770K 22nm Haswell, Intel® Core® i5-3570K 22nm Ivy Bridge, the AMD® FX™ family of processors, AMD® FX-4300, FX-6300, and FX-8350 32nm Vishera, AMD® Kaveri processors, Texas Instruments® Jacinto C6000™ automotive infotainment processors, Texas Instruments® OMAP™ automotive-grade mobile processors, ARM® Cortex™-M processors, ARM® Cortex-A and ARM926EJ-S™ processors, other industry-equivalent processors, and may perform computational functions using any known or future-developed standard, instruction set, libraries, and / or architecture.

[0100] Any of the steps, functions, and operations discussed herein can be performed continuously and automatically.

[0101] However, to avoid unnecessarily obscuring the present disclosure, the preceding description omits a number of known structures and devices. This omission is not to be construed as a limitation of the scope of the claimed disclosure. Specific details are set forth to provide an understanding of the present disclosure. It should however beappreciated that the present disclosure may be practiced in a variety of ways beyond the specific detail set forth herein.

[0102] Furthermore, while the exemplary embodiments illustrated herein show the various components of the system collocated, certain components of the system can be located remotely, at distant portions of a distributed network, such as a LAN and / or the Internet, or within a dedicated system. Thus, it should be appreciated, that the components of the system can be combined in to one or more devices or collocated on a particular node of a distributed network, such as an analog and / or digital telecommunications network, a packet-switch network, or a circuit-switched network. It will be appreciated from the preceding description, and for reasons of computational efficiency, that the components of the system can be arranged at any location within a distributed network of components without affecting the operation of the system. For example, the various components can be located in a switch such as a PBX and media server, gateway, in one or more communications devices, at one or more users’ premises, or some combination thereof. Similarly, one or more functional portions of the system could be distributed between a telecommunications device(s) and an associated computing device.

[0103] Furthermore, it should be appreciated that the various links connecting the elements can be wired or wireless links, or any combination thereof, or any other known or later developed element(s) that is capable of supplying and / or communicating data to and from the connected elements. These wired or wireless links can also be secure links and may be capable of communicating encrypted information. Transmission media used as links, for example, can be any suitable carrier for electrical signals, including coaxial cables, copper wire and fiber optics, and may take the form of acoustic or light waves, such as those generated during radio-wave and infra-red data communications.

[0104] Also, while the flowcharts have been discussed and illustrated in relation to a particular sequence of events, it should be appreciated that changes, additions, and omissions to this sequence can occur without materially affecting the operation of the disclosure.

[0105] A number of variations and modifications of the disclosure can be used. It would be possible to provide for some features of the disclosure without providing others.

[0106] In yet another embodiment, the systems and methods of this disclosure can be implemented in conjunction with a special purpose computer, a programmed microprocessor or microcontroller and peripheral integrated circuit element(s), an ASIC or other integrated circuit, a digital signal processor, a hard-wired electronic or logic circuitsuch as discrete element circuit, a programmable logic device or gate array such as PLD, PLA, FPGA, PAL, special purpose computer, any comparable means, or the like. In general, any device(s) or means capable of implementing the methodology illustrated herein can be used to implement the various aspects of this disclosure. Exemplary hardware that can be used for the present disclosure includes computers, handheld devices, telephones (e.g., cellular, Internet enabled, digital, analog, hybrids, and others), and other hardware known in the art. Some of these devices include processors (e.g., a single or multiple microprocessors), memory, nonvolatile storage, input devices, and output devices. Furthermore, alternative software implementations including, but not limited to, distributed processing or component / object distributed processing, parallel processing, or virtual machine processing can also be constructed to implement the methods described herein.

[0107] In yet another embodiment, the disclosed methods may be readily implemented in conjunction with software using object or object-oriented software development environments that provide portable source code that can be used on a variety of computer or workstation platforms. Alternatively, the disclosed system may be implemented partially or fully in hardware using standard logic circuits or VLSI design. Whether software or hardware is used to implement the systems in accordance with this disclosure is dependent on the speed and / or efficiency requirements of the system, the particular function, and the particular software or hardware systems or microprocessor or microcomputer systems being utilized.

[0108] In yet another embodiment, the disclosed methods may be partially implemented in software that can be stored on a storage medium, executed on programmed general- purpose computer with the cooperation of a controller and memory, a special purpose computer, a microprocessor, or the like. In these instances, the systems and methods of this disclosure can be implemented as program embedded on personal computer such as an applet, JAVA® or CGI script, as a resource residing on a server or computer workstation, as a routine embedded in a dedicated measurement system, system component, or the like. The system can also be implemented by physically incorporating the system and / or method into a software and / or hardware system.

[0109] Although the present disclosure describes components and functions implemented in the embodiments with reference to particular standards and protocols, the disclosure is not limited to such standards and protocols. Other similar standards and protocols not mentioned herein are in existence and are considered to be included in thepresent disclosure. Moreover, the standards and protocols mentioned herein and other similar standards and protocols not mentioned herein are periodically superseded by faster or more effective equivalents having essentially the same functions. Such replacement standards and protocols having the same functions are considered equivalents included in the present disclosure.

[0110] The present disclosure, in various embodiments, configurations, and aspects, includes components, methods, processes, systems and / or apparatus substantially as depicted and described herein, including various embodiments, sub combinations, and subsets thereof. Those of skill in the art will understand how to make and use the systems and methods disclosed herein after understanding the present disclosure. The present disclosure, in various embodiments, configurations, and aspects, includes providing devices and processes in the absence of items not depicted and / or described herein or in various embodiments, configurations, or aspects hereof, including in the absence of such items as may have been used in previous devices or processes, e.g., for improving performance, achieving ease and\or reducing cost of implementation.

[0111] The foregoing discussion of the disclosure has been presented for purposes of illustration and description. The foregoing is not intended to limit the disclosure to the form or forms disclosed herein. In the foregoing Detailed Description for example, various features of the disclosure are grouped together in one or more embodiments, configurations, or aspects for the purpose of streamlining the disclosure. The features of the embodiments, configurations, or aspects of the disclosure may be combined in alternate embodiments, configurations, or aspects other than those discussed above. This method of disclosure is not to be interpreted as reflecting an intention that the claimed disclosure requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment, configuration, or aspect. Thus, the following claims are hereby incorporated into this Detailed Description, with each claim standing on its own as a separate preferred embodiment of the disclosure.

[0112] Moreover, though the description of the disclosure has included description of one or more embodiments, configurations, or aspects and certain variations and modifications, other variations, combinations, and modifications are within the scope of the disclosure, e.g., as may be within the skill and knowledge of those in the art, after understanding the present disclosure. It is intended to obtain rights which include alternative embodiments, configurations, or aspects to the extent permitted, includingalternate, interchangeable and / or equivalent structures, functions, ranges, or steps to those claimed, whether or not such alternate, interchangeable and / or equivalent structures, functions, ranges, or steps are disclosed herein, and without intending to publicly dedicate any patentable subject matter.

Claims

CLAIMSWhat is claimed is:

1. A system comprising: a microprocessor; and a computer readable medium, coupled with the microprocessor and comprising microprocessor readable and executable instructions that, when executed by the microprocessor, cause the microprocessor to: analyze source code generated by an Artificial Intelligence (Al) algorithm, using a similarity algorithm, to identify a snippet of source code generated by the Al algorithm that is the same or similar to source code used to train the Al algorithm; determine if the identified snippet of source code generated by the Al algorithm is used in a software application; and generate a bill-of-materials for the software application that comprises information associated with the identified snippet of source code generated by the Al algorithm.

2. The system of claim 1, wherein the generation of the bill-of-materials for the software application is based on determining that the identified snippet of source code generated by the Al algorithm is used in the software application.

3. The system of claim 1, wherein the similarity algorithm uses at least one of the following to identify the snippet of the source code generated by the Al algorithm: a vector, a hash, and the source code used to train the Al algorithm.

4. The system of claim 1, wherein the bill-of-materials for the software application comprises information associated with other components used in the software application that are not generated by the Al algorithm.

5. The system of claim 1, wherein the bill-of-materials for the software application comprises one or more of: information associated with a specific version of the Al algorithm, information associated with a specification version of the source code generated by the Al algorithm, information associated with a specific version of the source code used to train the Al algorithm, information associated with a specific time that input data was used to generated the source code generated by the Al algorithm, and information associated with the input data used to generate the source code generated by the Al algorithm.

6. The system of claim 5, wherein the bill-of-materials for the software application indicates that each of the following is associated the identified snippet of source code generated by the Al algorithm: the specific version of the Al algorithm, the specification version of the source code generated by the Al algorithm, the specific version of the source code used to train the Al algorithm, and the input data used to generate the source code generated by the Al algorithm.

7. The system of claim 1, wherein the bill-of-materials for the software application comprises at least one of: information about the Al algorithm, license information about the source code used to train the Al algorithm, a likely license associated with the identified snippet of source code generated by the Al algorithm, trust information associated with the identified snippet of source code generated by the Al algorithm, a hash of the identified snippet of source code generated by the Al algorithm, the identified snippet of source code generated by the Al algorithm, a vector generated by the similarity algorithm, the source code generated by the Al algorithm, information associated with the software application, testing information, source code removal information, modification information, and release information.

8. The system of claim 7, wherein the bill-of-materials for the software application comprises the likely license associated with the identified snippet of source code generated by the Al algorithm.

9. The system of claim 7, wherein the bill-of-materials for the software application comprises the vector generated by the similarity algorithm.

10. The system of claim 7, wherein the bill-of-materials for the software application comprises at least one of: the hash of the identified snippet of source code generated by the Al algorithm and the identified snippet of source code generated by the Al algorithm.

11. The system of claim 1, wherein the bill-of-materials for the software application is stored in a blockchain that comprises one or more blocks associated with the bill- of-materials for the software application and wherein the one or more blocks associated with the bill-of-materials for the software application comprises at least one of: an Al algorithm block, a training code block, a license block, a likelylicense block, a trust information block, a hash block, a snippet block, a vector block, a user input block, an Al input block, a generated source code block, a software application block, an Al generated source code modification block, testing block, a release block, and a release lock block.

12. The system of claim 11, wherein the one or more blocks associated with the bill- of-materials comprises the release lock block and wherein the release lock block comprises a link that points back to a genesis block of the blockchain.

13. The system of claim 11, wherein the one or more blocks associated with the bill- of-materials comprises the likely license block and wherein the likely license block identifies a license associated with the identified snippet of source code generate by the Al algorithm.

14. The system of claim 11, wherein the one or more blocks associated with the bill- of-materials comprises the vector block.

15. The system of claim 11, wherein the one or more blocks comprises at least one of: the hash block that comprises a hash of the identified snippet of the source code generated by the Al algorithm and the snippet block that comprises snippet of the source code generated by the Al algorithm.

16. The system of claim 1, wherein the bill-of-materials for the software application comprises information associated with at least one of: two or more sets of Al generated source code from two or more different versions of the Al generated source code, two or more sets of Al generated source code generated at different times where nothing else changed other than time, two sets of Al generated source code with two different sets of source code used to train the Al algorithm, and modification information associated with the source code generated by the Al algorithm.

17. The system of claim 1, wherein the identified snippet of source code generated by the Al algorithm is based on a plurality of licenses associated with the source code used to train the Al algorithm and wherein the license information is shown as varying percentages.

18. A method comprising:analyzing, by a microprocessor, source code generated by an Artificial Intelligence (Al) algorithm, using a similarity algorithm, to identify a snippet of source code generated by the Al algorithm that is the same or similar to source code used to train the Al algorithm; determining, by the microprocessor, if the identified snippet of source code generated by the Al algorithm is used in a software application; and generating, by the microprocessor, a bill-of-materials for the software application that comprises information associated with the identified snippet of source code generated by the Al algorithm.

19. The method of claim 18, wherein the similarity algorithm uses at least one of the following to identify the snippet of the source code generated by the Al algorithm: a vector, a hash, and the source code used to train the Al algorithm.

20. The method of claim 18, wherein the bill-of-materials for the software application comprises one or more of information associated with a specific version of the Al algorithm, information associated with a specification version of the source code generated by the Al algorithm, information associated with a specific version of the source code used to train the Al algorithm, information associated with a specific time that input data was used to generated the source code generated by the Al algorithm, and information associated with the input data used to generate the source code generated by the Al algorithm.

21. The method of claim 18, wherein the bill-of-materials for the software application comprises at least one of information about the Al algorithm, license information about the source code used to train the Al algorithm, a likely license associated with the identified snippet of source code generated by the Al algorithm, trust information associated with the identified snippet of source code generated by the Al algorithm, a hash of the identified snippet of source code generated by the Al algorithm, the identified snippet of source code generated by the Al algorithm, a vector generated by the similarity algorithm, the source code generated by the Al algorithm, information associated with the software application, testing information, source code removal information, modification information, and release information.

22. The method of claim 21, wherein the bill-of-materials for the software application comprises the likely license associated with the identified snippet of source code generated by the Al algorithm.

23. The method of claim 21, wherein the bill-of-materials for the software application comprises the vector generated by the similarity algorithm.

24. The method of claim 18, wherein the bill-of-materials for the software application is stored in a blockchain that comprises one or more blocks associated with the bill-of-materials for the software application and wherein the one or more blocks associated with the bill-of-materials for the software application comprises at least one of: an Al algorithm block, a training code block, a license block, a likely license block, a trust information block, a hash block, a snippet block, a vector block, a user input block, an Al input block, a generated source code block, a software application block, an Al generated source code modification block, a testing block, a release block, and a release lock block.

25. The method of claim 24, wherein the one or more blocks associated with the bill- of-materials comprises the release lock block and wherein the release lock block comprises a link that points back to a genesis block of the blockchain.

26. The method of claim 24, wherein the one or more blocks associated with the bill- of-materials comprises the likely license block and wherein the likely license block identifies a license associated with the identified snippet of source code generate by the Al algorithm.

27. The method of claim 18, wherein the bill-of-materials for the software application comprises information associated with at least one of: two or more sets of Al generated source code from two or more different versions of the Al generated source code, two or more sets of Al generated source code generated at different times where nothing else changed other than time, two sets of Al generated source code with two different sets of source code used to train the Al algorithm, and modification information associated with the source code generated by the Al algorithm.

28. The method of claim 18, wherein the identified snippet of source code generated by the Al algorithm is based on a plurality of licenses associated with the sourcecode used to train the Al algorithm and wherein the license information is shown as varying percentages.

29. A non-transient computer readable medium having stored thereon instructions that cause a processor to execute a method, the method comprising instructions to: analyze source code generated by an Artificial Intelligence (Al) algorithm, using a similarity algorithm, to identify a snippet of source code generated by the Al algorithm that is the same or similar to source code used to train the Al algorithm; determine if the identified snippet of source code generated by the Al algorithm is used in a software application; and generate a bill-of-materials for the software application that comprises information associated with the identified snippet of source code generated by the Al algorithm.

Images