AI transparency notice

You're always told when you're talking to an AI

Every Future Assistants chat surface carries a disclosure banner — "You're chatting with an AI, not a human" — shown at the top of the conversation. This is a permanent, non-dismissible part of the chat UI, not a one-time onboarding notice.

AI output can be wrong

Future Assistants' AI features (assistant chat, onboarding, content generation) use AI models from third-party providers. AI output can be wrong, incomplete, or inappropriate, and should not be relied on without independent human verification — especially in these gated categories, per Terms Section 6:

• Health / medication — not offered at v1; assistants do not give medication or chronic-illness guidance.
• Dating — not offered at v1.
• Regulated financial advice — gated. Assistants refuse and recommend a qualified professional rather than answer directly.
• Regulated legal advice — gated, same behaviour: refuse and recommend a qualified professional.

AI Heart — EU AI Act Article 50 disclosure

AI Heart is a software layer that, with your explicit per-dimension consent (Settings → AI → Heart), infers aspects of your emotional and motivational state from your own messages across eight dimensions (desire, ambition, empathy, purpose, autonomy, competence, relatedness, self-acceptance). Full detail is in Privacy Policy Section 2.16.

Under the EU AI Act (Regulation (EU) 2024/1689) Article 50, we disclose this inference to you in advance, and it only begins after your explicit consent. We classify this processing as limited-risk. It is not deployed in workplace or education contexts, where emotion recognition is prohibited under Article 5(1)(f), and it does not exploit vulnerabilities in a way that materially distorts behaviour under Article 5(1)(b). Our anti-manipulation controls are documented in the AI Heart & Species Thesis.

AI Heart data is never used for pricing, ads, upsells, marketing, or retention prompts, and it is not visible to Future Assistants staff, third parties, advertisers, employers, or insurers without your per-incident permission. You can pause or delete any dimension at any time in Settings.

Safety-routing disclosure

Separately from AI Heart, if a message indicates self-harm, eating disorder, abuse, or severe-depression markers, the assistant refuses to continue in-character and routes you to a human support service, as described in Terms Section 6.1 and Privacy Policy Section 2.8. This is not therapy or medical care. If you are in immediate crisis, call your local emergency services (999 in the UK, 911 in the US, 112 in the EU).

The AI does not have a Heart

Assistants are explicitly instructed not to claim or imply they have feelings of their own. "AI Heart" reflects your emotional state back to you — it is not the assistant's feelings, and we do not anthropomorphise the AI.