Meta CEO Mark Zuckerberg said the company held back its Muse AI tool for months to harden safety and security, arguing labs should self-impose safeguards rather than accept an industrywide slowdown. His comments land amid a widening rift over the pace of frontier AI: Anthropic’s Dario Amodei and OpenAI’s Sam Altman have urged more caution and “pacing,” while Washington weighs new rules ranging from mandated kill switches to independent audits. Microsoft AI chief Mustafa Suleyman criticized Anthropic’s “constitutional” training approach, warning it could make models harder to control. Polls show rising concern about AI’s environmental footprint and data-center buildouts, and China is pushing for tighter international guardrails. The debate is increasingly geopolitical as U.S. officials prepare to host China’s Xi Jinping with AI high on the agenda.
Related article:
