AI kill switches could become reality after researchers flagged concerns: Big Technology's Kantrowitz

September 10, 2026 - 23:32

The idea of an emergency stop button for artificial intelligence is gaining traction. Alex Kantrowitz, founder of Big Technology, recently spoke about the growing debate over whether AI systems need built-in shutdown mechanisms. His comments came during a segment on Closing Bell, where he examined how the conversation around AI has shifted from practical concerns to speculative fiction.

Kantrowitz pointed out that the tech industry has done itself few favors when it comes to public trust. Years of overpromising and underdelivering have left many people skeptical of new breakthroughs. That skepticism now colors how regulators, workers, and everyday users view AI tools. When companies talk about safety, the audience often hears marketing rather than genuine caution.

The push for kill switches is not just a technical question. It raises hard issues about who gets to decide when to pull the plug. A model that helps diagnose diseases might also be used to spread misinformation. A system that writes code could also generate malicious software. Lawmakers in several countries are drafting rules that would require developers to include shutdown options before deployment.

Critics argue that kill switches could be abused. A government might silence an AI that exposes corruption. A corporation might disable a system that reveals its own failures. Still, the momentum is real. Researchers, ethicists, and even some industry insiders now say that without a reliable off switch, the risks of advanced AI may outweigh the benefits. The debate is no longer theoretical. It is moving from conference panels to legislative chambers, and the decisions made in the next few years will shape how powerful these systems become.