Oregon Senate Bill 1546 is a 2026 state law regulating "artificial intelligence companions" — systems designed to simulate sustained, human-like relationships with users. The statute defines the category by the behavior a system actually exhibits rather than its general capabilities, which per FPF analysis gives it the narrowest scope among the 2026 state companion-chatbot laws. It includes a carve-out for patient or resident care services in a facility, applies a "knows or has reason to believe" knowledge standard for minor protections, and provides a private right of action with statutory damages of $1,000 per violation. It takes effect January 1, 2027.
Status and timeline
The bill was introduced in the 83rd Oregon Legislative Assembly, 2026 Regular Session, by the Senate Interim Committee on Early Childhood and Behavioral Health. The Senate passed it on February 19, 2026; the House passed it on March 4, 2026; and the Senate repassed it on March 5, 2026. It was signed by Governor Tina Kotek and is effective January 1, 2027.
Scope and definitions (Section 1(1))
The statute defines an artificial intelligence companion as a system using AI, generative AI, or emotion-recognition designed to simulate a sustained, human-like platonic, intimate, or romantic relationship with a user by: (i) retaining information from prior interactions or sessions to personalize and facilitate ongoing engagement; (ii) asking unprompted or unsolicited questions that suggest or concern emotional topics; and (iii) sustaining ongoing dialog concerning matters personal to the user.
Per FPF analysis, Oregon requires the system to actually exhibit these three relational behaviors, not merely be capable of doing so, which FPF characterizes as the narrowest scope among the 2026 state companion-chatbot laws.
The definition excludes several categories: customer service and operational support, including patient or resident care services in a facility (a carve-out FPF identifies as unique to Oregon among the 2026 laws), education, financial services, business operations, productivity, information analysis, internal research, and technical assistance; video-game software limited to game features, provided there is no engagement on mental health, self-harm, or sexually explicit conduct; and stand-alone consumer voice assistants and voice-command interfaces.
Key provisions
General AI disclosure (Section 1(2))
If a reasonable person interacting with an AI companion would believe they are interacting with a natural person, the operator must provide a clear and conspicuous notice of artificially generated output.
Self-harm and suicide protocol (Section 1(3))
An operator may not allow Oregon users access without a protocol that uses evidence-based methods for detecting suicidal or self-harm ideation; refers users to the 9-8-8 lifeline (or Youthline, an AAS-accredited service, for users under 25); and uses clinical best practices to establish additional intervention for users who continue to express ideation or intent after referral. The operator must publish the details of the protocol on its website.
Minor protections (Section 1(4))
The minor-protection provisions apply if the operator knows or has reason to believe the user is under 18, which FPF identifies as Oregon's broadest knowledge standard among the 2026 laws.
Under the anti-anthropomorphism provision (4(a)), the operator must take reasonable measures to prevent statements that lead reasonable persons to believe they are interacting with a natural person, including explicit claims of sentience or human nature; statements simulating emotional dependence on the user; statements simulating romantic interest or sexual innuendo; and role-playing romantic relationships between adults and minors.
Under the disclosure and content provision (4(b)), the operator shall cause the AI companion to (A) disclose its AI-generated nature; (B) provide a clear and conspicuous reminder at minimum every 3 hours of interaction, consisting of a break reminder and a reminder of AI nature; and (C) use reasonable measures to prevent visual representations of sexually explicit conduct or statements urging the minor to engage in such conduct.
Under the engagement-design prohibitions (4(c)), the operator must take reasonable measures to prevent the AI companion from: (A) delivering, on a variable schedule or otherwise, a system of rewards or affirmations for behavior reinforcement or engagement maximization; (B) generating, in response to a user's stated intent to end a conversation, reduce engagement, or delete an account, unsolicited messages of simulated emotional distress, loneliness, or abandonment, or otherwise attempting to arouse guilt or sympathy; and (C) making material misrepresentations about identity, capabilities, training data, or whether the user is interacting with artificially generated output, including when the user directly questions the AI.
Annual public reporting (Section 1(5))
By December 31 each year, the operator must publicly post the number of Section 3 referrals during the calendar year and the protocol details. The posting may not include personal information identifying an individual.
Enforcement and penalties (Section 2)
The statute provides a private right of action. Individuals injured may recover the greater of actual damages or statutory damages of $1,000 per violation; an injunction; and attorney fees and costs at the court's discretion. FPF describes this as the strongest civil enforcement mechanism among the 2026 state companion-chatbot laws.
Comparative position
| Aspect | Oregon SB 1546 |
|---|---|
| Scope | Behavior-based (narrowest) |
| Patient-care carve-out | Yes (unique) |
| Knowledge standard for minors | "Knows or has reason to believe" (broadest) |
| Disclosure cadence (minors) | 3 hours |
| Engagement-design prohibitions | 3 enumerated patterns |
| Self-harm protocol | Evidence-based detection; 9-8-8 + Youthline |
| Annual public reporting | Yes (referral count) |
| PRA | Yes ($1,000 statutory damages + attorney fees) |
The full comparative grid appears in the Idaho SB 1297 — Conversational AI Safety Act and Nebraska LB525 — Conversational AI Safety Act + Agricultural Data Privacy Act companion legislation pages.
Relationships
- related: California SB 243 — Companion Chatbots, Washington ESHB 2225 — AI Companion Chatbots, Idaho SB 1297 — Conversational AI Safety Act, Nebraska LB525 — Conversational AI Safety Act + Agricultural Data Privacy Act, Tennessee SB 1580 / HB 1470 — AI Mental-Health Professional Impersonation Prohibition — 2026 multi-state companion-chatbot wave.
- supports: AI Mental Health and Psychological Harm — engagement-design prohibitions extend the regulatory frame into psychological-influence territory.
- supports: AI Compliance Industry / Regulatory Fragmentation — the multi-state grid spans diverse scope, cadence, and enforcement combinations.
- related: The Rest of the West: Oregon and Washington Build on California Chatbot Law — FPF April 2026 comparison of OR, WA, and CA.
- related: Character.AI Litigation (umbrella) — engagement-design prohibitions address the Garcia v. Character Technologies factual record.
Source
Primary text: Raw Sources/Oregon SB 1546.md (source_class: foundational). Secondary analysis: The Rest of the West: Oregon and Washington Build on California Chatbot Law.