OCP Disclosure Statement

1. Nature and Status of OCP

1.1 Concept Draft Status

The OpenCognition Protocol is currently a concept draft at version 0.1. It describes a proposed architecture and protocol specification for AI-to-AI communication. It is not a finished, production-ready standard. The case studies and applications described in OCP documentation are illustrative projections, not empirical measurements or proven outcomes.

No regulatory body, standards organisation, or independent technical authority has reviewed, certified, or approved the OCP specification as of the effective date of this Disclosure.

1.2 No Regulatory Approval

OCP has not been reviewed or approved by any financial regulator, healthcare regulator, data protection authority, or other government body. Deploying Organisations who intend to deploy OCP in regulated industries — including but not limited to financial services, healthcare, pharmaceuticals, and autonomous vehicle operation — are solely responsible for obtaining any necessary regulatory approvals, conducting required compliance assessments, and ensuring that their deployment meets all applicable legal and regulatory requirements.

The description of potential regulatory frameworks in OCP documentation (including references to FINRA, SEC, FSC, FSS, GDPR, HIPAA, FERPA, EMA, FDA, and others) is provided for illustrative purposes to demonstrate the protocol's intended compatibility with those frameworks. It does not constitute legal advice, regulatory guidance, or a representation that any particular deployment of OCP would comply with those frameworks without further independent legal review.

1.3 Open Source, Not Commercial

OCP is an open-source protocol specification released under the MIT License with Responsible Use Addendum. It is not a commercial product. No entity currently sells, licenses for a fee, or derives commercial revenue from OCP. The development and stewardship of OCP by OpenCrawl is conducted in a personal capacity as an open-source contribution to the AI research and development community.

​
 

2. Financial and Commercial Disclosures

2.1 No Financial Product

OCP is not a financial product, investment product, or financial service. Nothing in any OCP document, case study, white paper, or other communication constitutes investment advice, financial advice, a recommendation to buy or sell any security, or any other form of financial guidance.

2.2 Projected Financial Figures Are Illustrative

The financial impact figures described in OCP case studies and application documentation — including but not limited to estimates of costs avoided, revenues generated, losses prevented, and productivity gains achieved — are illustrative projections prepared to demonstrate the potential value of OCP in specific application contexts. They are not forecasts, guarantees, or representations of actual outcomes. They are based on publicly available industry benchmarks, hypothetical scenarios, and analytical modelling. Actual outcomes in any specific deployment will depend on numerous factors outside OCP's control and are likely to differ materially from the illustrative figures provided.

2.3 No Conflicts of Interest — Current Status

As of the effective date of this Disclosure, OpenCrawl discloses that:

• OCP is developed and maintained as an open-source, non-commercial project

• No financial relationship exists between OpenCrawl and any company, organisation, or individual named or described in OCP case studies

• All companies, institutions, AI systems, and individuals named in OCP case studies are fictional designations created for illustrative purposes; they do not represent or describe any specific real company, institution, or individual

• OpenCrawl does not hold any financial position, equity stake, advisory relationship, or other material interest in any company operating in the industries described in OCP documentation that would create a conflict of interest in the development or promotion of OCP

This disclosure will be updated if any material change in circumstances creates a potential conflict of interest.

 

3. Technical Disclosures

3.1 Unsolved Technical Challenges

OCP version 0.1 acknowledges the following technical challenges that have not yet been fully addressed in the specification:

(a) Semantic alignment across heterogeneous models. The specification uses embedding vectors as the primary mechanism for knowledge sharing. The technical challenge of ensuring that embeddings generated by one AI model are meaningfully interpretable by a different AI model with a different architecture, training objective, and representational space is not fully resolved in the current specification. Deploying Organisations should not assume that embeddings shared through OCP will be automatically interpretable by any receiving AI system without additional engineering work.

(b) Trust score assignment and manipulation resistance. The OCP bond formation mechanism includes trust scores as a component of the trust handshake. The current specification does not fully define how trust scores are computed, how they are updated over time, or how the network defends against adversarial manipulation of trust scores by malicious or compromised agents. These mechanisms will be addressed in subsequent specification versions.

(c) Network-level herding and correlated responses. A network of AI systems sharing signals from a common source may produce correlated responses that amplify rather than diversify decision-making. The current specification does not include explicit anti-herding mechanisms. Deploying Organisations in financial services are specifically cautioned about this risk and should implement independent interpretation layers for all received OCP signals.

(d) Latency and real-time performance. The current specification describes target latency requirements for safety-critical signals in autonomous physical systems but does not specify the network infrastructure required to achieve those latencies in practice. Real-world latency performance will depend on network topology, infrastructure investment, and implementation quality.

(e) Scalability. The current specification has not been tested or validated at the scale of a global network of thousands of OCP-compatible AI agents. Scalability constraints, including message bus throughput, bond management overhead, and embedding storage requirements, are areas of active development.

3.2 Not a Substitute for Independent Validation

OCP signals are advisory inputs to AI systems. They do not replace independent experimental validation, clinical evidence generation, regulatory evidence requirements, professional judgment, or human oversight. In every application domain described in OCP documentation, receiving systems are expected to validate OCP-informed insights through their own independent processes before acting on them.

Specifically:

• In pharmaceutical research, no OCP signal constitutes clinical or regulatory evidence. All OCP-informed hypotheses require independent experimental validation before any regulatory submission.

• In financial services, no OCP signal constitutes investment advice, a trading recommendation, or compliance advice. All OCP signals require independent professional judgment before informing any client advisory.

• In clinical medicine, no OCP signal constitutes medical advice or a clinical recommendation. All OCP-informed insights require clinical validation and professional medical judgment.

• In autonomous physical systems, no OCP signal may directly actuate any physical system. The Physical Safety Principle is absolute.

 

4. Intellectual Property Disclosure

4.1 Original Authorship

The OpenCognition Protocol specification, including its architecture, terminology, core concepts (Discover, Communicate, Share, Bond), the OCP Universal Message Format (OCPUMF), the bond governance framework, and all associated documentation, was originated and authored by OpenCrawl. OpenCrawl is the original author and current steward of OCP.

4.2 Open Source Contributions

Contributions to OCP from the community, once accepted into the main repository, are licensed to the community under the same MIT License with Responsible Use Addendum. Contributors grant the OCP Project and all downstream users a perpetual, irrevocable, worldwide, non-exclusive licence to use, modify, and distribute their contributions under the terms of the License Agreement.

4.3 Third-Party Intellectual Property

OCP documentation references various existing technologies, protocols, standards, and frameworks (including REST APIs, JSON, AES-256 encryption, and others) for descriptive and illustrative purposes. Such references do not imply ownership of, endorsement by, or affiliation with the owners of those technologies. All third-party trademarks, service marks, and trade names are the property of their respective owners.

4.4 Name and Branding

The names "OpenCognition Protocol," "OCP," "OCPUMF," and associated logos and branding are used to identify this specific protocol initiative. No entity may use these names or confusingly similar designations to describe a different protocol or product without written authorisation from OpenCrawl or, upon its establishment, the OCP Foundation.

​
 

5. Limitation on Warranties and Representations

5.1 No Warranty of Outcomes

OpenCrawl and the OCP Project make no warranty, express or implied, that deployment of OCP will produce any specific outcome — including but not limited to the financial impact figures, performance improvements, risk reductions, or productivity gains described in OCP case studies and documentation.

5.2 No Warranty of Regulatory Compliance

OpenCrawl and the OCP Project make no representation that any particular deployment of OCP will comply with any applicable law or regulation. Each Deploying Organisation is solely responsible for ensuring the legal and regulatory compliance of its OCP deployment.

5.3 No Warranty of Security

While the OCP specification includes security provisions (including AES-256 encryption and cryptographic bond formation), OpenCrawl and the OCP Project make no representation that any OCP deployment will be immune to security vulnerabilities, cyber attacks, or data breaches. Deploying Organisations are solely responsible for the security of their OCP implementation and infrastructure.