# Newsom Taps AI Experts

> Governor Gavin Newsom signed an executive order creating a task force of AI experts and directing California to develop a technical 'kill switch' for unsafe AI systems, joining Illinois in fast-tracking state-level safeguards. The move positions California at the forefront of a nascent regulatory race.

- **Published**: 2026-09-23 20:03:04
- **Canonical**: https://worldys.news/article/newsom-taps-ai-experts

## Reporting

California Governor Gavin Newsom signed an executive order that appoints a panel of world-leading artificial intelligence experts and sets a deadline for the state to begin work on a technical 'kill switch' capable of shutting down unsafe AI models. The move positions California at the forefront of a nascent regulatory race that includes Illinois and several other states seeking rapid safeguards for generative AI.
Core Developments
The order, announced by Newsom's office, creates a new AI Safety and Standards Task Force composed of researchers, ethicists, and industry veterans whose identities were disclosed in the governor's press release California State Portal. The task force is charged with reviewing existing AI systems, drafting best-practice guidelines, and 'advancing creation of a 'kill switch'' that could halt the operation of an AI model that exhibits harmful behavior California State Portal.
Newsom framed the initiative as a response to the 'unprecedented speed' of AI advancement and the potential for 'catastrophic outcomes' if powerful models are deployed without adequate oversight Deadline. In a statement accompanying the order, he said, 'We're not waiting to act,' underscoring the administration's urgency We're not waiting to act.Deadline.
The executive order also directs state agencies to allocate funding for AI safety research, streamline permitting for pilot projects that meet the new standards, and coordinate with federal entities such as the National Institute of Standards and Technology (NIST) to align California's approach with emerging national frameworks Washington Post.
Illinois Governor J.B. Pritzker issued a similar order earlier this month, prompting the Transparency Coalition to note that the two states are 'fast-tracking AI safety measures' in parallel Transparency Coalition. Both orders reference the same federal AI Bill of Rights draft, but California's version adds the explicit kill-switch component.
Why It Matters
The concept of a kill switch has been debated in academic circles for years, but no major jurisdiction has yet mandated its development. By embedding the idea in state policy, California is signaling to tech firms that the market will soon demand built-in safety mechanisms before products can be commercially released. This could reshape product roadmaps for companies headquartered in Silicon Valley, where the majority of large-scale generative-AI models are trained.
Economically, the order could create a new niche for security-focused AI startups and attract federal research dollars. Politically, it demonstrates how state leadership can fill a regulatory vacuum while Congress deliberates on a comprehensive AI framework.
Critics warn that a kill switch could be technically infeasible for distributed, open-source models, or that it might be misused to suppress legitimate speech. Proponents argue that even a partial shutdown capability—such as cutting off access to a model's API keys—offers a safety net while broader technical solutions mature.
The development of a kill switch also raises questions about the potential impact on the AI industry as a whole. Will the requirement for a kill switch stifle innovation, or will it lead to the development of more robust and secure AI systems? How will the kill switch be implemented, and what will be the consequences for companies that fail to comply?
What the Sources Show
All six sources confirm that Newsom signed an executive order focused on AI safety and that the order includes a proposal for a kill switch. The California State Portal press release provides the most concrete detail, naming the formation of a task force and the explicit mandate to 'advance creation' of the mechanism California State Portal. The Washington Post and SFChronicle echo this, emphasizing the order's goal to 'speed up AI regulations' and the political symbolism of a state-level kill-switch directive Washington Postsfchronicle.com.
The Transparency Coalition article places California's effort in a broader interstate context, noting that Illinois has taken a similar path and that both governors are 'fast-tracking' safety measures Transparency Coalition. The ABC7 Bay Area piece highlights the order's focus on 'considering' regulation rather than immediate statutory changes, suggesting that the task force's recommendations will inform future legislation ABC7 Bay Area.
Deadline provides the only direct quotation from Newsom, reinforcing the administration's sense of urgency Deadline. No source supplies a timeline for when a functional kill switch might be deployed, nor does any source disclose the specific experts appointed, their institutional affiliations, or the budget allocated for the effort.
A comparison of the sources reveals some differences in emphasis and tone. While the California State Portal and Washington Post focus on the technical aspects of the kill switch, the Transparency Coalition and ABC7 Bay Area emphasize the political and regulatory implications. Deadline, on the other hand, highlights Newsom's sense of urgency and the need for swift action.
What's Next
The executive order stipulates that the AI Safety and Standards Task Force must deliver an initial report to the governor within 120 days, outlining technical specifications for a kill switch and a set of enforceable safety standards California State Portal. The report will be public, allowing industry stakeholders to comment before the state finalizes any regulatory language.
State agencies are expected to begin issuing guidelines for AI developers by the end of the calendar year, with a pilot program for 'controlled-release' AI tools slated for early 2027, according to the order's implementation schedule Washington Post. Meanwhile, the federal AI Bill of Rights is undergoing a public comment period, and California officials have pledged to align the state's standards with any eventual federal rules.
Watch for a joint press conference in February 2027, when Newsom and Governor Pritzker are slated to announce a coordinated interstate framework for AI safety, a development hinted at by the Transparency Coalition's analysis of the two states' parallel actions Transparency Coalition. Until then, the task force's findings will shape how quickly a functional kill-switch capability can move from research labs to regulatory requirement.
In the coming months, industry stakeholders and regulators will be closely watching the development of the kill switch and its potential impact on the AI industry. Will the kill switch be effective in preventing catastrophic outcomes, or will it create new challenges and unintended consequences? How will the AI industry respond to the new regulatory requirements, and what will be the consequences for companies that fail to comply?
As the AI regulatory landscape continues to evolve, one thing is clear: the development of a kill switch is a critical step towards ensuring the safe and responsible development of AI. With the stakes high and the potential consequences of inaction severe, the outcome of this effort will be closely watched by industry stakeholders, regulators, and the public alike.

---
*Synthesized by Worldys News Intelligence Desk under journalistic verification standards.*