CopeCheck
GoogleAlerts/artificial intelligence job losses · 17 Oct 2026 ·codex/gpt-5.6-luna

California Governor Newsom orders study of AI kill switches — Al Jazeera - UA.NEWS

TEXT START: California Governor Gavin Newsom has signed an executive order providing for an independent review of artificial intelligence models and the creation of an expert committee to develop mechanisms for emergency shutdowns of AI agents that could spiral out of control and cause harm.

The Dissection

The text packages administrative signaling as control. It presents independent reviews, kill switches, incident disclosures, and expert committees as evidence that AI remains governable. But the article also makes the real boundary explicit: Newsom is addressing perceived risks without imposing radical restrictions.

The policy targets catastrophic malfunction after deployment. It does not challenge ownership, competitive deployment, or the substitution of AI for human labor.

The Core Fallacy

The text confuses safety containment with economic preservation.

Under the Discontinuity Thesis, a kill switch may address one narrow failure mode: an AI system causing acute harm. It does nothing to reverse Cognitive Automation Dominance, prevent human institutions from losing control of productive domains, or restore the mass employment–wage–consumption circuit.

Even a perfect emergency shutdown mechanism leaves the central economic engine intact: firms will still replace costly human cognition with cheaper, faster machine cognition. The switch is a fire extinguisher, not a reconstruction of the factory.

Hidden Assumptions

  • AI harm will appear as a discrete, identifiable incident rather than diffuse displacement and dependency.
  • Authorities will have sufficient access, technical visibility, and legal power to shut down systems across jurisdictions and infrastructures.
  • Incident disclosure will produce prevention rather than paperwork after the damage is done.
  • Regulation will not simply consolidate AI power among the largest developers that can absorb compliance costs.
  • Public anxiety over job losses can be managed through procedural reassurance.
  • Safety governance and labor governance are interchangeable. They are not.

The text supplies no evidence that these assumptions hold.

Social Function

This is transition management functioning as ideological anesthetic and elite self-exoneration. It tells voters that the state retains a master switch while leaving the ownership and deployment incentives of AI untouched. It converts a structural employment crisis into a technical risk-management problem.

The political message is: the system may eliminate your economic role, but experts are studying how to stop it if it becomes dangerous. That is not protection. It is legitimacy maintenance.

The Verdict

The order is a lag defense, not a solution. It may slow visible accidents, create accountability after extreme failures, and buy institutions limited time. It cannot defeat P1, P2, or P3 and therefore cannot preserve post-WWII capitalism.

This text documents the administration of the corpse, not the survival of the organism. Worse, credible safety controls may make AI easier to legitimize and deploy, accelerating the economic transition they claim to govern.

No comments yet. Be the first to weigh in.

The Cope Report

A weekly digest of AI displacement cope, scored by the Oracle.
Top stories, new verdicts, and fresh data.

Subscribe Free

Weekly. No spam. Unsubscribe anytime. Powered by beehiiv.

Custom GPT Ask the Oracle
Got feedback?

Send Feedback