'They're playing with our lives': AI researcher quits Anthropic

The departure of Jacob Coxon from Anthropic and the AI industry underscores a critical internal debate within leading US AI labs regarding the pace and safety of advanced model development. Coxon's claim that companies are "racing straight to self-improving superintelligence" without adequate safeguards points to a significant tension between innovation and risk management. This internal friction, publicly supported by Anthropic's own safety executive Evan Hubinger, who estimates a 10 percent risk of human extinction from AI within a decade, directly challenges the industry's narrative of controlled progress. For Asia, this US-centric debate has clear implications. As global AI development accelerates, the lack of federal regulation in the United States could push less cautious approaches to the forefront, potentially influencing standards and competitive pressures across Asian AI markets. The call for international coordination on AI development by OpenAI's chief scientist Jakub Pachocki suggests that regulatory frameworks and safety protocols established in one region will inevitably affect others. Asian governments and AI companies should monitor these internal US industry discussions closely, as they may foreshadow future global regulatory efforts or competitive shifts. The push for government intervention, as seen with Senator Bernie Sanders' bill to suspend AI development, reflects a growing recognition that self-regulation may be insufficient. If US policymakers move towards stricter oversight, it could create a precedent for how advanced AI is developed and deployed worldwide, including in major Asian AI hubs like China, South Korea, and Singapore. The core issue for Asia is whether it will adopt similar cautious stances or continue its own rapid development, potentially creating divergent regulatory landscapes.