This Addendum describes how Blunox uses data in connection with artificial-intelligence and machine-learning features of the Services ("AI Features") and forms part of the Agreement.
- No training on Customer Data (default). Blunox will not use Customer Data (including prompts, inputs, connected-system content, and outputs) to train, fine-tune, or improve any AI model made available to any other customer. Customer may separately opt in, in writing or via an explicit product setting, to permit specified uses; opt-in may be withdrawn prospectively at any time.
- Usage Data. Blunox may use aggregated or de-identified telemetry that does not identify Customer, Users, or any natural person to operate and improve the Services, including AI Features.
- Third-party models. Some AI Features are powered by third-party model providers listed on the subprocessor page (blunox.ai/legal/subprocessors). Blunox contracts with such providers on terms that (a) prohibit training on Customer Data, (b) impose confidentiality, and (c) meet the requirements of the DPA.
- Inputs and outputs. As between the parties, Customer owns its inputs and, to the extent permitted by law, the outputs generated for Customer by AI Features. Customer acknowledges outputs may not be unique across customers given similar inputs.
- Output disclaimer. AI Features produce probabilistic output that may be inaccurate or incomplete. Outputs are not professional advice. Customer is responsible for human review appropriate to its use, and for compliance with laws applicable to its use of AI Features (including AI-specific regulations such as the EU AI Act, where applicable to Customer as deployer).
- Restrictions. Customer will not use AI Features to develop competing models, to generate unlawful content, or in prohibited high-risk uses described in the AUP.
- Changes. Blunox will update this Addendum as AI Features evolve; material changes follow the amendment process in the General Terms.