NY, US · AI law tracker
A6767 — NY, US
A6767 is an AI governance legislation from NY, currently introduced. This bill introduces safety protocol and user notification mandates for AI companion models in New York [1]. AIGI tracks 1 primary-source update on this bill; the most recent was published on 2025-03-13.
Status & timeline
- Regulatory stage
- introduced
- Bill status
- Assembly Floor Calendar
- Authority / governing body
- New York State Senate
- Chamber
- Assembly
- Document type
- legislation
Next deadline: No fixed deadline — the bill is currently in legislative process.
Subscriber only
Full obligation matrix
| Actor | Obligation | Deadline | Source |
|---|---|---|---|
| provider | Implement a protocol within AI companion models to address possible suicidal ideation or self-harm expressed by a user. | — | — |
| provider | Implement a protocol within AI companion models to address possible physical harm to others expressed by a user. | — | — |
| provider | Implement a protocol within AI companion models to address possible financial harm to others expressed by a user. | — | — |
| provider | Provide notifications to users regarding crisis service providers. | — | — |
| provider | Provide notifications to users regarding the non-human nature of artificial intelligence companion models. | — | — |
Subscriber only
Enforcement risk score
Announced regulation; enforcement footprint still forming.
Subscriber only
Role-based compliance checklist
- product_manager Assess current AI companion models for capabilities to detect and respond to expressions of suicidal ideation, self-harm, physical harm to others, and financial harm to others.
- engineering Develop and integrate necessary protocols and features into AI companion models to address identified harms.
- compliance_officer Draft user notification language regarding crisis service providers and the non-human nature of AI models.
- general_counsel Review the bill's final text if enacted to ensure full legal compliance of new protocols and disclosures.
Subscriber only
Vendor impact assessment
- Vendor risk class
- high
- Procurement categories
- customer_service_ai, ai_agents
Vendors providing AI companion models will need to demonstrate robust internal protocols for detecting and addressing various forms of user-expressed harm and ensure mandatory user notifications about crisis services and the AI's nature.
Sample vendor questions
- What protocols does your AI companion model have in place to detect and respond to user expressions of suicidal ideation or self-harm?
- How does your AI system detect and address potential physical harm to others that a user might express?
- What mechanisms are in place to identify and mitigate risks of financial harm to others based on user interactions?
- How will your AI companion model ensure transparent notification to users about crisis service providers?
- How will your AI companion model explicitly inform users of its non-human nature?
Intelligence briefs (1)
New York Bill A6767 Mandates AI Companion Safety Protocols & Disclosures
This bill introduces safety protocol and user notification mandates for AI companion models in New York [1].
This development signals emerging deployer obligations for AI companion models, particularly concerning user safety and transparency.
Deadline: No fixed deadline — the bill is currently in legislative process.
Primary source →Frequently asked questions
- What is A6767?
- New York Bill A6767 (2025-2026) aims to regulate artificial intelligence companion models. The bill prohibits the provision of such models unless they incorporate protocols to address user expressions of suicidal ideation, self-harm, physical harm to others, or financial harm to others [1]. Additionally, it mandates specific user notifications regarding crisis service providers and the artificial intelligence companion's non-human nature [2]. Primary source →
- Why does A6767 matter?
- This development signals emerging deployer obligations for AI companion models, particularly concerning user safety and transparency. Primary source →
- Who does A6767 affect?
- This legislation primarily affects organizations developing, deploying, or offering artificial intelligence companion models to users within New York State. Considerations include any AI system designed for extended user interaction that might facilitate expressions of harm, requiring robust safety protocols and explicit user disclosures. Primary source →
- What are the key dates for A6767?
- No fixed deadline — the bill is currently in legislative process. Primary source →
- What is the current status of A6767?
- As of the last published update, A6767 is at the "introduced" stage, with bill status "Assembly Floor Calendar". Primary source →
- Where can I find the primary source for A6767?
- The primary source for the most recent update is at https://legislation.nysenate.gov/bills/2025/A6767. AIGI publishes the full citation chain plus every approved brief on this bill. Primary source →
Related
Stay informed