Former employees of Anthropic, OpenAI and Google DeepMind testified before the New York City Council about what they described as inadequate safeguards inside leading artificial-intelligence companies. Drawing on their industry experience, they warned that competitive pressure can encourage companies to release increasingly powerful systems before their risks are fully understood or independently evaluated. They urged city officials to require greater disclosure, outside safety audits and stronger protections for employees who report concerns. The hearing accompanied several proposed city measures addressing AI oversight, including requirements for transparency and independent review. The former employees’ testimony was especially significant because they used a formal public forum—and accepted potential professional consequences—to share firsthand concerns and advocate for enforceable safeguards rather than relying solely on voluntary corporate commitments.
Why It Matters
Artificial intelligence is becoming embedded in employment, education, government services, communications and other systems that affect people’s rights and opportunities. Yet much of the information needed to assess these technologies remains controlled by the companies developing them. Employees with direct knowledge may therefore be among the few people positioned to alert the public when internal safety practices appear insufficient. The story also demonstrates an important form of civic participation: technical experts leaving powerful companies and bringing their knowledge into a democratic decision-making process. Their testimony can help elected officials ask better questions and design more informed oversight. The witnesses’ call for transparency, independent scrutiny and protected whistleblowing addresses the public question of who should oversee technologies with potentially far-reaching consequences.