AI-Specific Disclaimer

1. Nature of AI-Generated Outputs

1.1 The Maveon platform enables users to deploy AI-powered conversational agents (“Chatbots”) that generate responses using large language models (LLMs) provided by third-party AI infrastructure providers including Anthropic (Claude), OpenAI, and Google.

1.2 AI-generated outputs are probabilistic and non-deterministic by nature. This means:

• The same input may produce different outputs at different times.
• Outputs are statistical predictions based on training data, not verified facts.
• AI models can and do produce responses that are factually incorrect, outdated, incomplete, ambiguous, inconsistent, or misleading — commonly referred to as “hallucinations.”
• No technical measure can guarantee the accuracy, reliability, or appropriateness of any AI-generated output.

1.3 Maveon does not review, verify, or validate any AI-generated output produced through a Client’s Chatbot. Maveon cannot and does not warrant that any output accurately reflects the training content, correctly answers a user’s question, or is suitable for any particular purpose.

2. No Professional Advice

2.1 Nothing generated by a Maveon-powered Chatbot constitutes, and must not be presented as:

• Legal advice — Chatbot outputs are not a substitute for advice from a qualified lawyer and do not create a solicitor-client relationship.
• Medical or health advice — Chatbot outputs are not a substitute for professional medical diagnosis, treatment, or advice from a licensed healthcare provider.
• Financial or investment advice — Chatbot outputs are not a substitute for advice from a registered financial adviser and do not constitute a recommendation to buy, sell, or hold any financial product.
• Tax advice — Chatbot outputs are not a substitute for advice from a qualified tax professional.
• Engineering, safety, or technical certification — Chatbot outputs must not be relied upon for safety-critical decisions.

2.2 Clients deploying Chatbots in any domain touching upon the above categories must ensure that their Chatbot includes clear and prominent disclosures directing end users to seek qualified professional advice.

3. Third-Party AI Model Disclaimer

3.1 The AI models powering Maveon Chatbots are developed and operated by third parties:

• Anthropic PBC — Claude language models
• OpenAI, LLC — GPT language models
• Google LLC — Gemini and associated AI services

3.2 Maveon has no control over the underlying model architecture, training data, safety alignment, or output filtering of these third-party models. Changes to these models by their providers — including changes to behavior, content policies, or output quality — may affect Chatbot outputs without prior notice to Maveon or to Clients.

3.3 Maveon disclaims all liability for any output, behavior, or characteristic of any third-party AI model, including outputs that are biased, discriminatory, harmful, or otherwise objectionable.

4. Client Validation Obligation

4.1 Before deploying any Chatbot to end users, Clients are solely and exclusively responsible for:

• Thoroughly testing Chatbot outputs across a representative range of anticipated user queries.
• Verifying that outputs are accurate, appropriate, and consistent with the Client’s content and brand standards.
• Implementing human review processes for high-stakes outputs where incorrect information could cause harm.
• Continuously monitoring deployed Chatbots for unexpected, harmful, or inaccurate outputs and taking prompt corrective action.
• Configuring appropriate system prompts, guardrails, and escalation paths to limit the scope of Chatbot responses.

4.2 Maveon expressly disclaims any liability for harm arising from a Client’s failure to validate Chatbot outputs prior to deployment or to monitor Chatbot behavior following deployment.

5. End User Disclosure Requirement

5.1 Clients must ensure that all end users interacting with a Maveon-powered Chatbot are clearly informed that they are communicating with an artificial intelligence system and not a human representative. This disclosure must be:

• Visible at or before the commencement of any conversation.
• Written in plain language.
• Not obscured by design or interface choices.

5.2 A Chatbot must not be configured to deny being an AI when sincerely and directly asked by an end user.

6. Limitation of Maveon’s Liability for AI Outputs

6.1 To the maximum extent permitted by law, Maveon has zero liability for any harm, loss, damage, expense, claim, or legal proceeding arising from or related to any AI-generated output, including outputs that are:

• Factually incorrect, misleading, or harmful;
• Defamatory, discriminatory, or offensive;
• In breach of any third-party right;
• In violation of any applicable law or regulation; or
• Otherwise inappropriate for the context in which they are presented.

6.2 The Client assumes all risk and liability for AI-generated outputs produced through their Chatbot and presented to their end users.