Law / United States / 
Pennsylvania
S.B. 1090, SAFECHAT Act
Pa. S.B. 1090, 2025-2026 Regular Session
Before the second chamber, dated , as of .
An AI transparency rule binding private bodies.
- Enforcement body
- Pennsylvania Office of Attorney General
- How the hook was established
- express
- What makes it apply
- Market targeting
- Obligation class
- Disclosure, Governance
- Audit expectation
- none
Where it has got to
The text described here is Printer's Number 1423, as amended on second consideration in the Senate. That print is PN 1423, published .
Locally, this stage is House Communications and Technology Committee.
The stage above is recorded at www.palegis.us.
Senate Communications and Technology Committee reported SB 1090 as committed 11-0 on . Amended on second consideration (PN 1423) , re-referred to Appropriations , re-reported as committed 22-0 and passed the Senate 49-1 (third consideration and final passage) on . Referred to the House Communications and Technology Committee ; no action recorded since. The 2025-2026 session runs through .
What it requires
- Issue a clear and conspicuous notification that an AI companion is artificially generated and not human, whenever a reasonable person interacting with it could otherwise be misled into believing it is human.NIST AI RMFMEASURE 2.8
- Maintain, implement, and publicly publish a protocol to prevent the AI companion from producing suicidal ideation, suicide, or self-harm content, including referring a user who expresses such content to a crisis line.NIST AI RMFMEASURE 2.6NIST AI RMFGOVERN 4.2
- For a user you know or should have known is a minor, give the additional disclosures and safeguards this measure specifies.NIST AI RMFMAP 1.6
These lines are LexLint's reading of the law's text.
In the frameworks
- NIST AI RMF
- GOVERN 4.2Organizational teams document the risks and potential impacts of the AI technology they design, develop, deploy, evaluate, and use, and they communicate about the impacts more broadly.
- MAP 1.6System requirements (e.g., “the system shall respect the privacy of its users”) are elicited from and understood by relevant AI actors. Design decisions take socio-technical implications into account to address AI risks.
- MEASURE 2.6The AI system is evaluated regularly for safety risks – as identified in the MAP function. The AI system to be deployed is demonstrated to be safe, its residual negative risk does not exceed the risk tolerance, and it can fail safely, particularly if made to operate beyond its knowledge limits. Safety metrics reflect system reliability and robustness, real-time monitoring, and response times for AI system failures.
- MEASURE 2.8Risks associated with transparency and accountability – as identified in the MAP function – are examined and documented.
- NIST AI 600-1
- GAI-RISK-07Human-AI Configuration
- GAI-RISK-08Information Integrity
- MIT mitigations Draft
- FINOS AIGF
- AIR-DET-021Agent Decision Audit and Explainability
- ETSI EN 304 223
Our reading of where this law's duties sit in each framework, and in the NIST AI RMF, line by line beside each duty above. It never says that running a control meets the law. How we read a framework against the law
Identifiers refer to ETSI EN 304 223 V2.1.1 (2025-12), © ETSI 2025, all rights reserved. ETSI owns the standard; no ETSI text (titles or sentences) is reproduced here. The standard is free to download from ETSI.
If you get it wrong
Criminal exposureNo
Penalty structure
As passed by the Senate; collected in a civil action by the Attorney General.
- Rule
- Per violation only
- As of
- Currency
- USD
- Per violation unit
- Violation
- Per violation amount
- 10,000
What it makes you log
Logging duty
Section 3 would require an AI companion operator to give a clear notification that the companion is not human, to maintain, implement and publish a protocol to prevent suicidal-ideation and self-harm content (with a crisis-service referral), and, for a user it knows or should have known is a minor, to disclose, remind the user to take a break every three hours, and take reasonable measures against sexually explicit output. The protocol is a safeguard design duty; the bill requires no log or record of a notification, a detection or an interaction, and no report to anyone.
- Kind
- None
- As of
What this law does
This measure passed the Pennsylvania Senate 49-1 on and was referred to the House Communications and Technology Committee on , and the bill status page records no report from that committee as of .
As passed by the Senate, an operator of an AI companion platform must issue a clear and conspicuous notification that the AI companion is artificially generated and not human whenever a reasonable person interacting with it could be misled into believing otherwise; must maintain and publish a protocol, to the extent technologically feasible, to prevent the AI companion from producing suicidal ideation, suicide, or self-harm content, including a notification referring a user who expresses such content to a crisis line; and, for a user the operator knows or should have known is a minor, must give additional disclosures and safeguards.
The bill exempts an underlying AI model not directly offered as a companion, and several narrow categories including business-internal tools and video-game bots limited to game topics. The Attorney General has exclusive enforcement authority, with a civil penalty of up to $10,000 per violation; the bill states no private right of action. The bill provides that it takes effect in 120 days.
When LexLint raises it
When your app profile says your app generates content with AI, serves under-18s or deploys a companion chatbot.
Back to the example · Lint your app
Page updated .