Abstract

AAM 2.0 marks the next stage in the evolution of AAM: from a summary measure of AI involvement in music to a structured AI contribution layer. Rather than producing only a single aggregate result around a Sound Recording, AAM 2.0 returns to two long-established domains in the music industry - the Musical Work and the Sound Recording - and describes AI participation in each separately.

This evolution is grounded in the operational experience of AAM 1.0. Since the release of ISRC-AAM-CID 1.0 in 2025, the framework has been applied to AI-music registration, discovery, presentation and market-facing use, producing a dataset of more than 1,600 registered songs. That experience suggests that the question "Was AI used?" is only the outermost layer. The industry increasingly needs to know which music entity AI contributed to, which creative role was affected, to what degree, who made the declaration, and whether provenance information exists that can support it.

AAM 2.0 therefore develops two distinct but interoperable modules. The first, Rights-Domain Contribution Measurement, separates the Musical Work from the Sound Recording and produces AAM-Work and AAM-Recording. The second, Provenance-Assisted Declaration, reads and validates available C2PA Content Credentials at the declaration stage, using provenance as a trigger and an additional source of assurance rather than as a substitute for the declaration itself.

The boundaries of the framework are equally important. ISWC continues to identify the Musical Work and ISRC continues to identify the Sound Recording; AAM does not replace either identifier. C2PA does not calculate AAM and does not establish a role-level AAM percentage. AAM does not determine authorship, the subsistence of copyright, infringement, or royalty entitlement. Its function is narrower: to express AI contribution in a structured, measurable and machine-readable form so that creators, labels, distributors, DSPs, rights organisations and other industry systems can read, exchange and use the information.

Within this architecture, AAM 2.0 can operate as an independent AI Contribution Layer within creator, distributor and DSP workflows, with relevant information mapped into DDEX metadata or AI-related advisory information for downstream exchange.

1. From AAM 1.0 to AAM 2.0: What Has Changed?

1.1 AAM 1.0: Turning AI Involvement into Structured Data

The original purpose of AAM 1.0 was to create a measurable information layer between free-text disclosure and a simple "AI / non-AI" label. It used a set of representative creative roles - Composer, Lyricist, Music Producer, Lead Performer / Vocalist, Instrumentalists, Recording Engineer, Mixing Engineer and Mastering Engineer - and combined role weights with AI Participation to translate complex human-AI collaboration into a recordable and interpretable contribution model.

The reference relationship was:

Role AI Contributionᵢ = Role Weightᵢ × AI Participationᵢ

In the AAM 1.0 reference model, the eight role weights total 100%, while 0%, 50% and 100% represent different levels of AI Participation. The example below produces an aggregate result of 42.5/100. AAM0-AAM4 can then map the continuous result into semantic bands for easier interpretation and exchange.

This model moved AAM beyond the binary question of whether AI was used and toward a more useful question: how much did AI participate in the creation of the music? It also provided the basis for a more granular contribution model.

1.2 Why a Single Overall Result Is Not Enough

The aggregate AAM 1.0 result is useful as a summary of overall AI involvement, but operational experience revealed an important limitation: one number cannot show whether AI contribution occurred in the Musical Work, the Sound Recording, or both.

The existing music industry already distinguishes these two entities. Lyrics, melody and composition belong to the Musical Work; performance, instrumentation, production, recording, mixing and mastering contribute to a particular Sound Recording. A single Musical Work may have multiple Sound Recordings. The industry correspondingly uses ISWC and ISRC to identify the two entities and has developed distinct rights, licensing and revenue structures around them.

Generative AI changes the visibility of that traditional process. A system may take prompts, lyrics, melodies, audio material or other inputs and participate in composition, lyric writing, vocal generation, instrumentation and production within the same technical environment, ultimately producing a near-finished recording. The output appears as one Sound Recording, but the creative contribution embodied in it can still span both the Musical Work and the Sound Recording.

This creates a structural problem for AI-era music data: when multiple stages that were previously visible in the creative process are compressed into a final recording, how can the contribution attributable to the Musical Work and the Sound Recording be made visible again, and how can Human Contribution and AI Contribution be described within each domain?

This issue was also reflected in discussion at the DDEX Nashville AI Working Group in September 2026. Participants noted that the AAM approach presented from Australia considered signals such as lyrics and therefore reached beyond the final recording alone. The discussion also raised the value of connecting Work-level and Recording-level AI information in a way that remains intelligible as a whole. For a listener, music is commonly experienced as a single object; for industry data exchange, rights management and licensing, the underlying entity and contribution domain still matter.

AAM 2.0 therefore needs to do two things at once: preserve the established identity and rights boundaries between the Musical Work and the Sound Recording, while providing an AI contribution information layer that can connect them.

This distinction is particularly important for Work-side rights management. If AI disclosure is attached only to the final Sound Recording, contribution at the composition and lyric level can be subsumed within recording-level information. A distinct Work-level expression allows songwriters, publishers, collective management organisations and other Work-side participants to receive clearer, exchangeable AI contribution information.

1.3 Two Independent but Connected Contribution Domains

AAM 2.0 therefore expands the single aggregate structure of AAM 1.0 into two contribution domains: one for the Musical Work and one for the Sound Recording.

AAM 1.0 placed composition, lyrics, production, vocals, instrumentation, recording, mixing and mastering within one 100-point structure. That was effective for producing an overall measure of AI involvement, but it also placed Work-side and Recording-side contribution in the same calculation space. Two tracks can therefore produce the same aggregate AAM result while having fundamentally different structures: one may use AI primarily for composition and lyrics, while another may have human-written music and lyrics but extensive AI participation in vocals, instrumentation and production. Those structures are materially different for Work-side and Recording-side rights.

AAM 2.0 treats each domain as its own complete 100% contribution base and produces two primary outputs:

AAM-Work and AAM-Recording are not isolated scores. They are structured results associated with different music entities within the same AAM framework. Measuring the two domains separately makes it possible to locate AI contribution; keeping them connected within the same framework preserves the ability to disclose, exchange and present a piece of music as a coherent whole.

This dual-domain approach does not redefine the Musical Work or the Sound Recording, nor does it alter ISWC, ISRC or the rights systems built around them. It addresses a new data problem created by generative AI: when stages previously distributed across Work creation and Recording production are compressed into a final output, their contribution structure needs to be re-expressed before AI participation can be meaningfully measured.

AAM itself does not determine the economic value of either domain and does not establish authorship, copyright ownership or royalty entitlement. Its role is to identify where AI contribution occurs, which creative roles are involved, and how that participation is expressed in a structured form. Legal and economic treatment remains a matter for copyright, contract, licensing and market mechanisms.

The move from one aggregate result to AAM-Work and AAM-Recording is therefore more than a change in calculation. It is the foundation for AAM to evolve from AI-involvement disclosure toward a rights-aware AI contribution infrastructure.

2. The AAM 2.0 Technical Model: Rights Domain × Creative Role × AI Participation

Chapter 1 established what AAM 2.0 needs to measure: AI contribution should be expressed separately for the Musical Work and the Sound Recording. This chapter defines how that measurement is formed.

AAM 2.0 retains the role-based logic of AAM 1.0 but introduces Rights Domain as the first layer of the model. A contribution is therefore no longer associated only with a creative role; it is first associated with the Musical Work or the Sound Recording and then with a specific Creative Role and level of AI Participation.

2.1 Two Rights Domains

The eight reference roles from AAM 1.0 are reorganised according to the Musical Work and the Sound Recording. The Musical Work covers Work-side creative contribution and may be associated with an ISWC; the Sound Recording covers performance, production and technical contribution and may be associated with an ISRC.

The core AAM 2.0 data structure can therefore be expressed as:

Rights Domain × Creative Role × Relative Contribution × AI Participation

Rights Domain identifies the music entity to which the contribution relates. Creative Role identifies the creative function. Relative Contribution describes that role's share of contribution within its domain. AI Participation describes the degree to which AI participated in that role. The latter two variables together produce the AI Contribution for the role.

2.2 Independent Normalisation Within Each Domain

AAM 2.0 treats the Musical Work and the Sound Recording as separate contribution domains. Within each domain, the Relative Contribution of all Creative Roles sums to 100%.

Σ Role Weightᵢ = 100%, i ∈ Work

Σ Role Weightᵢ = 100%, i ∈ Recording

For any role i within a domain:

AI Contributionᵢ = Role Weightᵢ × AI Participationᵢ

For either domain D:

AAM-Domain = Σ (Role Weightᵢ × AI Participationᵢ), i ∈ D

AAM-Work and AAM-Recording therefore both range from 0 to 100%, but each measures AI contribution within its own domain. They are not a value ratio between the Musical Work and the Sound Recording.

This preserves the principle established in Chapter 1. The domains are separated so that the location of AI contribution can be identified, while independent normalisation creates a consistent measurement base within each domain. The two 100% bases are measurement structures, not an allocation of economic value.

2.3 Reference Weight Conversion from AAM 1.0

AAM 2.0 does not require the role weights and operational data developed under AAM 1.0 to be discarded. To preserve continuity, the eight AAM 1.0 reference weights can be used as an initial reference profile and renormalised separately within the Musical Work and Sound Recording domains.

This conversion is intended primarily to demonstrate mathematical continuity between AAM 1.0 and AAM 2.0. The resulting weights are a reference profile, not a claim that AAM 2.0 has fixed them as permanent industry weights.

Musical Work

Under AAM 1.0, Composer and Lyricist carried weights of 20 and 15 respectively, for a combined total of 35. Renormalising those roles within the Musical Work produces:

AAM-Work = 50 / 100

Under this reference profile and declaration, 50% of the contribution within the Musical Work is measured as AI participation.

Sound Recording

The six AAM 1.0 roles associated with the Sound Recording carried a combined original weight of 65. Applying the same domain-normalisation method produces:

AAM-Recording ≈ 38.46 / 100

The same AAM 1.0 example can therefore be re-expressed as AAM-Work = 50% and AAM-Recording ≈ 38.46%. Compared with the original aggregate result of 42.5/100, the significance of AAM 2.0 is not merely numerical: it preserves the music-entity and rights-domain context of the AI contribution.

2.4 Core Outputs of AAM 2.0

The primary outputs of AAM 2.0 are AAM-Work and AAM-Recording. Together they provide domain-level AI contribution information for a piece of music.

In the reference example, AAM-Work is 50/100 and AAM-Recording is approximately 38.46/100. If the AAM 1.0 global weighting structure is retained for summary purposes, AAM-Overall remains 42.5/100.

These values have different meanings. AAM-Work and AAM-Recording retain rights-domain information and are the principal structured outputs of AAM 2.0. AAM-Overall may remain useful for historical compatibility, a single summary indicator or simplified consumer-facing presentation, but it should not replace the two domain-level measurements.

2.5 Standardised AI Participation Inputs

Relative Contribution and AI Participation are distinct variables. Relative Contribution describes the share of a Creative Role within its rights domain; AI Participation describes the degree to which AI participated in that role's contribution.

To preserve usability and continuity with AAM 1.0, the initial AAM 2.0 model can retain 0%, 50% and 100% as standard reference inputs:

The 0%, 50% and 100% values are standardised reference values for declaration and calculation. They do not imply that real-world human-AI collaboration exists in only three states. As declaration data, production records and provenance information become richer, AAM may support more granular AI Participation while maintaining compatibility with the base semantic levels.

This variable must also remain distinct from C2PA-assisted provenance. AI Participation is an AAM declaration and measurement input. Provenance information may support or strengthen that declaration, but it does not automatically generate an AI Participation percentage.

2.6 Default Profile and Contextual Contribution Profile

The actual contribution of a Creative Role can vary substantially between projects. A permanently fixed set of weights improves comparability but may fail to reflect different creative and production models. Fully custom weights for every project, by contrast, would weaken consistency and cross-record comparability.

AAM 2.0 therefore proposes two contribution profiles:

The Default AAM Profile is intended for cases where more specific contribution evidence is unavailable or where a common baseline is needed. It should remain public, versioned and relatively stable. A Contextual Contribution Profile allows role weights to be adjusted where production records, collaboration agreements, session data or other reasonable supporting information provide a basis for doing so.

Both profiles use the same AAM calculation logic; the principal difference is the source of Relative Contribution. In either case, Role Weight describes relative contribution within the AAM model. It does not represent contractual ownership, copyright share or royalty split.

3. C2PA + AAM: Provenance-Assisted Declaration

Chapter 2 defines how AAM 2.0 measures contribution. A separate question remains: where do the declarations used in that calculation come from, and is there provenance information that can support them?

AAM 2.0 therefore introduces Provenance-Assisted Declaration alongside contribution measurement. The central principle is to keep declaration and provenance as independent information layers while allowing validated provenance to trigger, inform and strengthen an AAM declaration. C2PA-compatible Content Credentials are an important provenance source for this purpose.

3.1 The Role of C2PA in AAM 2.0

C2PA provides machine-readable information about the provenance and processing history of digital content. Depending on implementation and what upstream tools write into the asset, Content Credentials may include information about creation, editing, tools used, processing actions and related assertions.

AAM addresses a different question. It describes whether AI contribution relates to the Musical Work or the Sound Recording, identifies the relevant Creative Role, records the degree of AI Participation, and calculates AAM-Work and AAM-Recording.

The boundary can be stated simply:

C2PA establishes provenance. AAM establishes contribution structure.

C2PA therefore does not enter the mathematical formula for AAM-Work or AAM-Recording and is not a second AI-contribution score. Instead, it functions as an auxiliary source layer around the AAM declaration, allowing a record to capture not only what was declared but, where available, what provenance information can support or contextualise that declaration.

3.2 Declaration and Provenance Remain Independent Sources

AAM 2.0 treats Creator / Rightsholder Declaration and C2PA-Compatible Provenance as two separate information sources.

Keeping the sources independent is essential. A creator declaration is a structured claim about a particular music-creation process. Provenance records information that can be read and validated from the digital asset or its provenance chain. The two may support one another, but they are not the same information and neither should be treated as a substitute for the other.

For example, Content Credentials may indicate that a generative tool participated in creating or processing an asset. That fact alone does not establish Composer AI Participation at 50%, Vocalist at 100%, or Mixing Engineer at 0%. Role-level contribution data still needs to be expressed through the AAM declaration and calculated under the contribution model in Chapter 2.

Conversely, an AAM declaration of AI participation does not imply that C2PA provenance must exist for the asset. The source, status and verification scope of each information layer should remain visible.

3.3 Provenance-Assisted Declaration Rules

At the declaration stage, AAM 2.0 may read and validate available C2PA Content Credentials and adapt the subsequent workflow to the result.

The purpose is not to use provenance to decide whether a piece of music "used AI". It is to make use of machine-readable provenance that already exists, reduce information gaps, and trigger a more complete AAM declaration where appropriate.

This creates an important boundary: absence of provenance is not evidence of absence of AI use.

The presence of C2PA provenance depends on whether upstream tools support the mechanism, whether Content Credentials are written, and whether those credentials survive later editing, export and distribution. A failure to find verified AI-related provenance therefore describes only the information state of the asset being examined; it cannot be converted into a conclusion that AI was not used.

Likewise, the presence of AI-related provenance establishes only that the provenance contains AI-related source or processing information. Its meaning still needs to be interpreted in light of the specific assertions, validation status and AAM declaration.

3.4 Two Principal Functions of Provenance in AAM

Within AAM 2.0, provenance information serves two principal functions: Declaration Trigger and Assurance Enhancement.

As a Declaration Trigger, validated provenance related to generative AI can prompt or require the declarant to complete the relevant Musical Work, Sound Recording and Creative Role information. Existing machine-readable provenance can therefore become an entry point into structured music disclosure without the system attempting to infer the full contribution structure on its own.

As Assurance Enhancement, where the AAM declaration can be associated with tools, creation events or other assertions recorded in provenance, the relevant provenance reference can be retained as supporting information. Downstream systems can then distinguish a declaration-only record from one that also carries provenance support.

The verification scope must remain explicit. C2PA can validate provenance within the scope of its mechanism, but this should not be expressed as "C2PA verified the AAM score". AAM-Work and AAM-Recording are calculated from AAM declaration data; provenance supports independently verifiable source information, not the entire contribution result.

This distinction is also important for any future AAM assurance mechanism: verifying the source of information is not the same as verifying every contribution claim.

3.5 AAM 2.0 Provenance-Assisted Workflow

In implementation, Provenance-Assisted Declaration can operate as a distinct stage within AAM registration or data ingestion. The system first identifies the Musical Work and Sound Recording and any available identifiers, while also reading available provenance. An authorised declarant then completes the relevant role-level contribution declaration, after which the AAM 2.0 model produces AAM-Work and AAM-Recording.

This workflow does not require every AAM record to contain C2PA provenance. An AAM declaration can stand independently, with provenance added where available. Equally, the existence of C2PA Content Credentials does not mean that a complete AAM declaration can be generated automatically, because Rights Domain, Creative Role, Relative Contribution and AI Participation remain AAM-defined information.

Provenance-Assisted Declaration therefore does not merge C2PA and AAM into a single verification system. It creates a defined interface between them: C2PA provides verifiable provenance information, while AAM converts AI contribution relevant to music creation and rights domains into structured data and, where possible, associates the two within the same record.

4. From Contribution Measurement to Industry Exchange: DDEX-Compatible Interoperability

The value of AAM 2.0 does not end with the calculation of AAM-Work and AAM-Recording. For structured AI contribution information to become useful in the music industry, it must be capable of being recognised and exchanged across creators, rightsholders, labels, publishers, distributors, DSPs and rights organisations.

AAM 2.0 therefore does not seek to create a new music-data exchange system. It is designed as an AI Contribution Information Layer that can connect with the existing music supply chain. AAM defines and measures AI contribution; established exchange frameworks such as DDEX provide the interoperability through which relevant information can move between systems and business participants.

4.1 AAM as an Exchangeable AI Contribution Information Layer

The music supply chain already contains mature data structures around Musical Works, Sound Recordings, Releases, Parties, rights and business information. AAM 2.0 is not intended to alter those entities or require the industry to adopt a new composite identifier that tightly binds ISWC, ISRC and AAM. Its purpose is to add a structured information layer describing AI contribution within the existing architecture.

ISWC therefore continues to identify the Musical Work and ISRC continues to identify the Sound Recording. AAM-Work and AAM-Recording are associated with the relevant entities to describe the degree of AI participation within each rights domain.

This approach allows AAM information to be referenced and exchanged at different levels depending on the use case. A receiving system may consume domain-level measurements such as AAM-Work or AAM-Recording and, where necessary, access more granular Creative Role, AI Participation, declaration-source or provenance information without changing the semantics or lifecycle of existing identifiers.

The interoperability objective of AAM 2.0 is therefore not to transmit a new "AAM code" as an end in itself. It is to make AI contribution a structured information object that different music systems can understand and exchange.

4.2 Interoperability with DDEX

DDEX provides standardised data-exchange frameworks for participants across the digital music supply chain. The relationship between AAM 2.0 and DDEX is therefore not one of replacement or a parallel exchange standard. The practical question is how AAM contribution information can be mapped into existing or evolving industry data structures.

The two layers perform different functions. AAM 2.0 defines the structure, semantics and measurement of AI contribution in the Musical Work and Sound Recording. DDEX-compatible interoperability addresses which parts of that information need to enter a real sender-to-receiver workflow and whether they are best carried as existing fields, extended data or external references.

AAM does not require DDEX to absorb the entire internal AAM data model. In a real supply chain, some AAM information may be suitable for direct inclusion in industry messages, while more granular information may be better accessed through an external reference or API. That boundary should be validated against the relevant DDEX standards, use cases and actual sender / receiver requirements.

This approach also allows the AAM model to evolve independently. Creative Roles, AI Participation semantics, provenance handling and contribution profiles can develop over time without changing established identifiers or requiring every downstream system to ingest the full AAM structure.

4.3 How Different Industry Participants May Use AAM Information

The industry value of separating the Musical Work from the Sound Recording becomes clearer once AAM information enters the supply chain. Different participants do not need the same level or type of AI information; they can use the layer that corresponds to their rights domain and business function.

This differentiated use is one of the practical reasons for separating the Musical Work and Sound Recording in AAM 2.0.

If AI information is attached only to the final Sound Recording, composition- and lyric-level contribution is difficult to pass independently to publishers, collective management organisations or other Work-side participants. AAM-Work provides a structured AI contribution object directly relevant to the Work domain, while AAM-Recording provides the corresponding information for labels, master rightsholders and other Recording-side participants.

The same granularity does not need to be exposed to consumers. The underlying layer can preserve Work-level, Recording-level and role-level information while a consumer-facing or use-case-specific application presents a simpler disclosure or advisory signal. Granular infrastructure and simple presentation are not in conflict.

5. Next Steps

AAM 2.0 remains a Discussion White Paper. The next phase should focus not on expanding the framework indefinitely, but on testing the core changes proposed here: dual-domain measurement for the Musical Work and Sound Recording, provenance-assisted declaration, and interoperability with the existing music supply chain.

OZBeat AI will first test AAM-Work and AAM-Recording within existing AI-music registration workflows. Building on AAM 1.0 data and registration experience, this work will examine the Work / Recording role structure, the Default Profile, AI Participation inputs and the model's applicability across different music-creation scenarios.

A second area of testing will be the practical availability of C2PA-compatible Content Credentials in real music assets. The focus will be on what provenance information different generative-AI tools and production workflows actually provide, and how that information can trigger, inform or strengthen AAM declarations without replacing creator or rightsholder disclosure.

For industry interoperability, the next phase will examine concrete sender-to-receiver use cases and map AAM-Work, AAM-Recording and relevant disclosure information to DDEX exchange structures. The objective is not to embed the entire AAM model inside a single message standard, but to determine which information needs to travel through the existing music supply chain and which information is better accessed as an external structured reference.

AAM 2.0 should also continue to be tested with creators, songwriters, publishers, labels, distributors, DSPs, rights organisations, AI companies, provenance practitioners and standards communities. A particular priority is to validate the real information needs of Work-side and Recording-side participants and the boundaries of use in copyright, licensing and revenue-management contexts.

These steps should be driven by operational data, real workflows and industry feedback. Future versions of AAM 2.0 should evolve from those results rather than attempting to settle every technical, rights and economic question in advance.

Conclusion: From AI Participation to AI Contribution

Generative AI is changing how music is created. It is also changing how the music industry can observe creative contribution.

Historically, music could be traced through comparatively visible creative and production processes into a Musical Work and a Sound Recording, and then into Work-side rights, Recording-side rights, licensing and revenue systems. Generative AI can compress composition, lyrics, vocals, instrumentation and production into the same technical environment and produce a near-finished recording directly. As a result, contribution relationships that were once visible across a creative chain can become difficult to observe, while simple labels such as "AI-generated" or "AI-assisted" become increasingly inadequate for industry use.

AAM 1.0 addressed the first part of this problem by converting AI involvement from a binary label into structured data organised around Creative Roles. AAM 2.0 takes the next step by placing that measurement back inside the music industry's established identity and rights structure. AAM-Work and AAM-Recording allow AI contribution in the Musical Work and Sound Recording to be described independently while remaining connected within one framework.

C2PA-compatible provenance can add source information to the declaration process; DDEX-compatible interoperability can carry relevant contribution information into the existing music supply chain.

AAM 2.0 deliberately approaches the problem from the Output Side of generative AI. It does not attempt to identify every work or recording used in model training, nor does it replace training attribution, licensing or other input-side rights systems. Its focus is narrower and more tractable: once new music has been created with human and AI participation, how can the contribution structure of the Musical Work and Sound Recording be made visible again, and how can Human Contribution and AI Contribution become measurable and exchangeable?

That focus also defines the limits of AAM 2.0. It does not redefine ISWC or ISRC. It does not determine authorship, copyright subsistence or infringement. It does not equate provenance with contribution verification, and it does not decide how much any participant should ultimately be paid. It provides a layer of information that those later mechanisms can use: where AI participated, in which music entity and creative role, and how that participation can be expressed in a structured measurement.

As AI music moves further into mainstream creation, distribution and consumption, the industry will need to address more than the question "Is this AI music?" It will need a way to understand new contribution relationships within the music rights structures that already exist. AAM 2.0 is intended as a framework that can be tested and refined toward that purpose - turning AI contribution from an opaque technical fact into information that the music industry can understand, exchange and eventually use in rights and economic systems.

Related Reading

• ISRC-AAM-CID 2.0: Towards an Interoperable Global Trust Infrastructure for AI Music
• OZBeat AI Presents AI Music Disclosure Framework at DDEX's First In-Person AI Working Group Meeting
• AI Assisted Music Registration Upgrade: Customisable Role Weights and Clearer AI Involvement Records
• ARIA Sets New Rules for AI Music: Why Disclosure Is Becoming Part of Music Infrastructure
• Spotify's AI Credits Are Live — And the Industry Is Heading Toward Granular AI Disclosure
• OZBeat AI Becomes a C2PA Contributor Member

Special Notice

AAM and the related ISRC-AAM-CID framework are developed and advanced by OZBeat AI. Except where third-party material is expressly identified, the original frameworks, models, text, tables and related content in this Discussion White Paper are owned by OZBeat AI. All rights are reserved.

This document is intended to support industry discussion, research, product design and technical validation of AAM 2.0. It does not constitute legal, copyright, tax, investment or financial advice. References to ISWC, ISRC, DDEX, C2PA or any other third-party standard, organisation or system are included solely to explain interoperability context and potential areas of technical alignment. They do not imply endorsement, approval or sponsorship of AAM 2.0, this White Paper or OZBeat AI by any third party.