Responsible AI Use Policy

Last Updated:: July 18, 2026
Effective Date:: July 18, 2026

Responsible AI Use Policy

Last Updated: July 18, 2026 Effective Date: July 18, 2026

This Responsible AI Use Policy (this "AI Policy") describes how My Opus, Inc. ("Opus," "Company," "we," "us," or "our") uses artificial intelligence and machine learning systems in connection with the Opus services (the "Services"), and the principles that guide our development, deployment, and operation of those systems.

This AI Policy supplements, and forms part of, our Terms of Use. Capitalized terms used but not defined in this AI Policy have the meanings given to them in our Terms of Use or our Privacy Policy, as applicable.

1. Purpose and Scope

The Services use artificial intelligence and machine learning systems (collectively, "AI Systems") to deliver personalized coaching content, generate Outputs, and operate the memory and personalization features of the Services. Because the responsible development and operation of AI is a matter of public interest, of regulatory focus, and of trust between Opus and its Users, we publish this AI Policy to describe our approach.

This AI Policy applies to all uses of AI Systems within the Services. It is intended to be read alongside our Terms of Use (in particular, Section 10 (Artificial Intelligence and Automated Processing) and Section 11 (Coaching Disclaimer; Not Professional Services)), our Privacy Policy (in particular, Section 5 (How AI Processing Works at Opus)), and our Acceptable Use provisions.

Nothing in this AI Policy modifies the legal terms set forth in our Terms of Use, our Privacy Policy, or our other policies. In the event of any conflict between this AI Policy and the Terms of Use or Privacy Policy, the Terms of Use or Privacy Policy (as applicable) will control.

2. Our AI Principles

We develop, deploy, and operate AI Systems in accordance with the following principles. These principles describe our intentions and our internal commitments; they are not, and should not be construed as, warranties or guarantees regarding the performance of AI Systems or Outputs.

2.1 Purposeful Use

We use AI only where it meaningfully improves the User experience or supports the operation, security, or improvement of the Services. We do not use AI for the sake of using AI.

2.2 Honest Communication

We describe what our AI Systems do and do not do in plain, accurate language. We do not claim that our AI Systems offer therapeutic, medical, psychological, or other professional services. We do not claim accuracy, reliability, or capabilities that the AI Systems do not have.

2.3 User Agency

We design the Services to position AI Outputs as inputs to your reflection, not as authoritative answers. Decisions belong to you.

2.4 Privacy by Default

We minimize the Personal Information sent to AI Systems, contractually limit how third-party AI providers may use that information, and prevent third-party providers from training their general-purpose models on User Content, as described in Section 5 (How AI Processing Works at Opus) of our Privacy Policy.

2.5 Continuous Improvement

We monitor AI System behavior, evaluate Outputs against quality and safety criteria, and update prompts, models, and methodologies as we learn from operation and feedback.

2.6 Accountability

We maintain internal records of meaningful AI System decisions (including model selection, prompt design, and significant changes to AI behavior) and we evaluate AI System performance and risks on a periodic basis.

3. What Our AI Does and Does Not Do

We describe below what our AI Systems do and do not do, in order to set accurate expectations and to support your informed use of the Services.

3.1 What Our AI Does

(a) Generates personalized coaching content based on your inputs and interactions, including prompts, lessons, quests, reflections, and summaries;

(b) Personalizes your experience by drawing on your User Content and on AI-generated representations (such as embeddings) of your past inputs to retrieve relevant context and tailor future Outputs;

(c) Determines, suggests, and refines a personality type designation and cognitive function profile under the cognitive function model developed by John Beebe. The initial type designation is based on your onboarding inputs, and may be updated, refined, or revised over time as you provide additional inputs and as our systems develop a more accurate representation of your cognitive function preferences. Our AI may suggest a revised type designation based on continued interaction; the type designation is exploratory and educational and is not, and should not be relied upon as, a diagnostic assessment;

(d) Surfaces themes and patterns that may be relevant to your goals and reflections;

(e) Supports operational functions of the Services, such as content categorization, search, and quality monitoring; and

(f) Where you elect to use social or community features, generates analyses comparing your type designation and cognitive function profile to the type designation of another User who has also elected to participate, for the purpose of providing observations regarding potential compatibility, friction, or communication patterns. Where both you and the other User have provided the necessary consent, such analyses may include specific examples drawn from your respective inputs; in the absence of such consent, analyses are limited to generalized observations based on type designation alone.

3.2 What Our AI Does Not Do

(a) Our AI does not provide medical, psychiatric, psychological, therapeutic, counseling, financial, legal, or other professional advice. Outputs are educational and reflective in nature only.

(b) Our AI does not diagnose, treat, prevent, or monitor any medical or mental health condition. Opus is not a medical device and is not used as a medical device.

(c) Our AI does not constitute a substitute for professional care or judgment. If you need professional support, you should seek the advice of a qualified, licensed professional.

(d) Our AI does not monitor User Content for signs of distress, crisis, or risk of harm in real time, and it should not be relied upon to detect or respond to emergencies. If you are experiencing a mental health crisis, please contact emergency services or a crisis hotline as set forth in Section 11.4 (Crisis Resources) of our Terms of Use.

(e) Our AI does not produce factually verified or authoritative information. Outputs may be inaccurate, biased, outdated, or misleading. You should evaluate Outputs critically and exercise your own judgment.

(f) Our AI does not make solely automated decisions that produce legal effects concerning you or that similarly significantly affect you within the meaning of Article 22 of the EU General Data Protection Regulation. Outputs are informational and reflective only.

(g) Our AI does not generate Outputs that are unique to you across all Users. Similar or identical Outputs may be generated for different Users.

(h) Our AI is not, and is not held out as, an autonomous agent acting on your behalf. It does not take actions outside of the Services without your direction.

4. How Our AI Systems Work

4.1 Components

Our AI Systems comprise:

(a) Large language models provided by third-party AI service providers, including (as of the Effective Date) Anthropic, accessed through commercial application programming interfaces under enterprise data processing agreements;

(b) Embedding models provided by third-party AI service providers, including (as of the Effective Date) OpenAI, accessed through commercial application programming interfaces under enterprise data processing agreements;

(c) Proprietary prompts, instructions, content libraries, and methodologies developed by Opus and applied to the third-party models above; and

(d) Memory, retrieval, and orchestration systems developed by Opus that select, sequence, and contextualize content for delivery to you.

4.2 Processing Flow

In general, when you interact with the Services, our AI Systems may:

(a) receive your input (for example, a journal entry, a response to a prompt, or a quest report);

(b) generate or update an embedding of your input for storage in our memory and retrieval systems;

(d) transmit your input, the retrieved context, and our proprietary prompts to a third-party AI service provider for processing under contractual confidentiality and data processing obligations;

(e) receive a generated Output from the third-party AI service provider;

(f) post-process the Output (for example, to apply formatting, safety checks, or contextual adjustments); and

(g) deliver the Output to you within the Services.

4.3 Personalization Over Time

Our AI Systems personalize your experience over time on an individual basis, based on your User Content, your stated goals, your reported experiences, and your engagement patterns. This personalization is specific to your account.

We may use aggregated, de-identified, or statistical data derived from User Content and interactions across all Users to evaluate, refine, and improve our prompts, our methodologies, our AI Systems, and the Services generally, as further described in Section 5.3 (Use of Aggregated Data to Improve the Services) of our Privacy Policy.

4.4 Subprocessors

Information regarding the third-party AI service providers we engage as Subprocessors is included in our Subprocessor List.

5. Limitations of AI Outputs

5.1 Probabilistic Nature

AI Outputs are produced through probabilistic processes. Outputs may be inaccurate, incomplete, inconsistent, biased, outdated, misleading, or otherwise problematic. AI Systems may also generate content that appears confident or authoritative but is factually incorrect (commonly referred to as "hallucinations"). You should evaluate Outputs critically and not rely on them as authoritative or accurate without independent verification.

5.2 Not Professional Advice

Outputs are not professional advice of any kind. See Section 11 (Coaching Disclaimer; Not Professional Services) of our Terms of Use.

5.3 No Real-Time Monitoring or Crisis Response

Our AI Systems do not monitor User Content for signs of distress, crisis, or risk of harm in real time, and should not be relied upon for emergency response. See Section 11.4 (Crisis Resources) of our Terms of Use.

5.4 Bias and Fairness

AI Systems can reflect or amplify biases present in training data, prompts, or the data they receive at runtime. While we work to identify and mitigate bias in our AI Systems, we cannot guarantee that Outputs will be free of bias. If you encounter Outputs you believe are biased, inappropriate, or harmful, please report them to support@opuswithin.com.

5.5 Evolving Capabilities

The capabilities and behaviors of our AI Systems evolve over time as we update prompts, content, models, and methodologies. We do not guarantee consistency of Outputs across versions or over time.

6. Safety, Content Moderation, and Abuse Prevention

6.1 Safety Mechanisms

We employ a combination of mechanisms intended to promote safe operation of our AI Systems, which may include:

(a) prompt engineering and system instructions designed to keep Outputs within scope and aligned with the educational and reflective purpose of the Services;

(b) content filters and safety classifiers (provided by us, by our third-party AI service providers, or both) designed to limit the generation of certain categories of harmful, unsafe, or off-topic content;

(d) human review and quality assurance processes for prompt design, content libraries, and AI System changes.

These mechanisms are not infallible, and may not detect or prevent all problematic Outputs.

6.2 Reporting Concerns

If you encounter Outputs you believe are unsafe, harmful, harassing, biased, factually incorrect in a way that could cause harm, or otherwise inconsistent with this AI Policy, please report them to support@opuswithin.com. We will review reports and take appropriate action, which may include adjustments to prompts or content libraries, escalation to our third-party AI service providers, or other remedial steps.

6.3 Prompt Injection and Adversarial Inputs

We are aware that AI Systems may be susceptible to "prompt injection" and other adversarial inputs designed to cause the AI Systems to produce unintended Outputs or to disclose information they should not. We take reasonable measures designed to detect and mitigate prompt injection. Use of the Services to attempt prompt injection or other adversarial manipulation is prohibited under Section 7 (Prohibited Uses) and Section 7 (Acceptable Use) of our Terms of Use.

6.4 Abuse Prevention

We may use AI Systems and other tools to help detect, prevent, and respond to abuse of the Services, including violations of our Terms of Use. We may take action against accounts that are determined to have engaged in abuse, including suspension, termination, and referral to law enforcement where appropriate.

7. Prohibited Uses

In addition to the prohibitions set forth in Section 7 (Acceptable Use) of our Terms of Use, you may not, and may not attempt to:

(a) use the Services or any AI Outputs to develop, train, fine-tune, evaluate, benchmark, or improve any artificial intelligence model, machine learning system, or competing product or service;

(b) attempt to extract, reverse engineer, or otherwise discover the prompts, system instructions, model weights, training data, or non-public methodologies of our AI Systems;

(c) submit to the Services any input designed to cause the AI Systems to produce harmful, unsafe, illegal, or otherwise prohibited Outputs (including by means of prompt injection, jailbreaking, or other adversarial techniques);

(d) use the Services or any AI Outputs to provide medical, psychological, psychiatric, therapeutic, counseling, financial, or legal advice to any person, or to imply that such advice has been provided;

(e) use the Services or any AI Outputs in connection with the development, design, training, or deployment of weapons, surveillance systems, biometric identification systems, or other systems intended to cause serious physical harm or to facilitate unlawful surveillance;

(f) use the Services or any AI Outputs to generate content that defames, harasses, threatens, or incites violence against any individual or group, or that violates the privacy or rights of any third party;

(g) use the Services or any AI Outputs to make decisions that produce legal effects or similarly significantly affect any individual (including decisions regarding employment, credit, housing, education, insurance, healthcare, or access to government services); or

(h) use the Services or any AI Outputs to generate content intended to deceive, defraud, or impersonate any person or entity.

We may, in our sole discretion, take appropriate action in response to any actual or suspected violation of this Section 7, including the actions described in Section 16 (Suspension and Termination) of our Terms of Use.

8. Human Oversight

We do not review every Output before it is delivered to you. The volume of Outputs generated by the Services makes individual human review infeasible, and individual human review would in any event compromise the personal nature of your interactions with the Services.

We do, however, exercise human oversight of our AI Systems through:

(a) human design, review, and approval of prompts, content libraries, methodologies, and significant AI System changes;

(b) human review of samples of Outputs (where appropriate, in accordance with our Privacy Policy) to evaluate quality and safety;

(d) human decision-making with respect to actions taken in response to violations of our Terms of Use, including suspension or termination of accounts.

You retain ultimate responsibility for evaluating and using Outputs. See Section 11 (Coaching Disclaimer; Not Professional Services) of our Terms of Use.

9. Transparency

9.1 What We Disclose

We disclose:

(a) the categories of AI Systems we use, in this AI Policy and in our Privacy Policy;

(b) the categories of third-party AI service providers we engage, in our Privacy Policy and our Subprocessor List;

(d) material changes to our AI practices, through updates to this AI Policy, our Privacy Policy, and (where appropriate) in-Service notices.

9.2 What We Do Not Disclose

We do not disclose, and we treat as confidential and proprietary, the specific text of our prompts, the design of our content libraries, the specific configurations of our AI Systems, or other competitively sensitive details. Disclosure of such information would compromise the security and competitive position of the Services and could enable abuse.

10. Your Rights and Choices Regarding AI Processing

10.1 Use of the Services Requires AI Processing

Use of the Services requires processing of your User Content by AI Systems. If you do not wish to have your information processed by AI Systems, you should not use the Services. We do not currently offer a non-AI version of the Services.

10.2 Privacy Rights

You retain all rights under our Privacy Policy and applicable law, including rights of access, correction, deletion, and objection. To exercise these rights, see Section 13 (Your Privacy Rights) of our Privacy Policy.

10.3 Account Deletion

You may delete your account at any time. Upon deletion, your User Content (including embeddings derived from your User Content) will be removed from our active systems in accordance with the retention schedule set forth in Section 9 (Data Retention) of our Privacy Policy.

11. Changes to This Policy

We may update this AI Policy from time to time. The "Last Updated" date at the top of this Policy indicates when it was last revised. Material changes will be communicated as set forth in Section 23 (Changes to These Terms) of our Terms of Use.

12. Contact Us

For questions about this AI Policy, please contact us at:

My Opus, Inc. Attention: Legal Department Email: legal@opuswithin.com Privacy: privacy@opuswithin.com Support: support@opuswithin.com