The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →California’s AI safety framework is now more than a proposal: the state enacted SB 53 in 2025, then signed SB 813 and AB 1405 in September 2026. Together, the measures establish public safety-framework disclosures and critical-incident reporting for covered frontier AI developers, alongside independent verification and a state registry and standards for AI auditors. A September 2026 executive order calls for recommendations on additional safeguards—including a possible independently checked “kill switch”—but those proposals are not themselves enacted legal duties.
What California has enacted
The measures are a developing framework, not one all-purpose AI law. SB 53 sets duties for covered frontier AI developers. SB 813 and AB 1405 add infrastructure for independent assessment and auditors. The state’s public announcements describe these measures at a high level; they do not, by themselves, establish every statutory definition, compliance date, or enforcement detail.
| Measure | Status and focus | What the available description says |
|---|---|---|
| SB 53 (2025), Transparency in Frontier Artificial Intelligence Act | Enacted in 2025 | Covered frontier AI developers must publicly disclose safety frameworks, report specified critical safety incidents to the state, and provide whistleblower protections for workers reporting serious risks. The Governor’s signing statement describes these duties; detailed coverage and compliance rules should be checked in the statutory text. |
| SB 813 (2026) | Signed in September 2026 | Establishes a framework for independent verification organizations to assess AI systems and models for compliance with state law and for safety and risk. The announcement does not specify a compliance date or penalty. |
| AB 1405 (2026) | Signed in September 2026 | Creates a state registry for AI auditors and standards addressing their independence, transparency, and integrity. The announcement does not specify a compliance date or penalty. |
The Governor’s 2025 signing statement said SB 53 requires disclosures, incident reporting, and whistleblower protections. The Associated Press also reported a public cloud for researchers and a $1 million per-violation fine. Those reported details should not be treated as a universal penalty or cloud entitlement without checking the enacted text and the law’s scope.
Who is covered—and what is not yet clear
SB 53 is directed at covered frontier AI developers, not every business that uses an AI product. The descriptions available from the Governor and news coverage do not supply the statutory thresholds or enough detail to determine which particular developer or system qualifies. SB 813 concerns verification organizations assessing AI systems and models; AB 1405 concerns auditors and their registration and standards. Exact definitions, exceptions, deadlines, and the relationship among the three laws require consultation of the enacted statutes and implementation guidance.
#1 Best Overall
That distinction matters for companies trying to understand their position: being an AI user is not, on the information summarized here, enough to conclude that a company has SB 53 developer duties. Nor does the creation of auditor or verification infrastructure alone establish that every AI system must undergo a particular audit.
What the September 2026 executive order adds—and does not add
On September 18, 2026, Governor Gavin Newsom issued an executive order directing agencies to speed implementation of SB 813 and AB 1405 and convene experts to recommend whether California should take further steps. The order identifies possible measures for consideration:
- Require onsite independent verifiers.
- Verify safety-framework disclosures and risk assessments.
- Require an independently checked frontier-model “kill switch.”
- Expand the definition of critical incidents.
These are recommendations or possible future changes, not requirements made operative merely by appearing in the order. A company should distinguish the direction to develop recommendations from an enacted statute or a binding implementation rule.
Why some technology companies opposed SB 53
The Associated Press reported that some technology companies opposed SB 53 and argued AI legislation should be handled at the federal level. The concern is that differing state requirements could produce a regulatory patchwork. That is not evidence that all technology companies opposed the law: Anthropic described the measures as practical safeguards formalizing practices many companies already followed voluntarily, while also supporting federal standards.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Rank #3
Anthropic co-founder and head of policy Jack Clark told AP: “While federal standards remain essential to avoid a patchwork of state regulations, California has created a strong framework that balances public safety with continued innovation.”
The dispute followed a negotiation over how far regulation should go. Newsom vetoed a broader predecessor in 2024 after concerns that its requirements were too rigid and could hamper innovation. He then convened experts, and SB 53 incorporated recommendations and industry feedback, according to AP. In his 2025 signing statement, Newsom said: “California has proven that we can establish regulations to protect our communities while also ensuring that the growing AI industry continues to thrive. This legislation strikes that balance.”
Rank #4
A separate proposed initiative is not the same as enacted law
A different proposal would create a California AI Safety Commission and regulate certain large AI companies. The Secretary of State’s 2026 notice said proponents needed to submit 546,651 signatures—five percent of votes cast for governor in November 2022—by August 3, 2026, for ballot eligibility. That notice records the threshold and deadline; it does not establish whether the initiative later qualified. Its status should not be described as enacted law or as a ballot measure without a newer official status record.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What businesses should verify before acting
The announcements establish the framework’s direction but do not resolve every operational question. Before treating a duty as applicable, a company should check the current statutory text and state implementation materials for:
- Whether its role and systems fit the law’s definitions of a covered frontier AI developer, system, auditor, or verification organization.
- Which safety-framework disclosures and critical incidents are covered, and when reporting or disclosure is due.
- How independent verification organizations and registered auditors are defined and what standards or procedures apply.
- Any effective dates, regulations, exceptions, enforcement provisions, or changes adopted after the September 2026 announcements.
This article describes the California measures at a policy level, based on official announcements and AP reporting available through October 4, 2026; it is not legal advice.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
