The call for a slowdown in the race to build powerful AI is losing support. Anthropic CEO Dario Amodei has argued for greater caution, but Meta CEO Mark Zuckerberg has rejected the idea of a coordinated pause. He believes AI companies should compete while independently building safety into their systems.
Zuckerberg says every AI company already has its own reason to make models safe. He points to Meta’s Muse AI agent, which he says was held back for months for safety checks. In his view, safe and reliable AI will be a product advantage. He also says that an AI system that ignores instructions will not find many users.
The problem is that the major players are not moving together. Some technology leaders support caution, while Meta, Nvidia’s Jensen Huang, Donald Trump and China favour faster development. A slowdown can work only when everyone agrees to follow it.
So, the AI race is likely to continue. The real challenge is to create common rules for testing, monitoring and controlling powerful systems before they cause harm. If governments and companies avoid that responsibility, the race may become faster than our ability to manage it.
THE AI RACE HAS BEGUN — BUT THE SAFETY RULES ARE STILL MISSING.
Sanjay Sahay
Have a nice evening.