- Meta CEO Mark Zuckerberg opposed a coordinated slowdown in advanced artificial intelligence development, saying each laboratory should set its own pace.
- Meta delayed the launch of its Muse AI agent by several months to strengthen its security and protection measures.
- Anthropic CEO Dario Amodei has proposed independent audits and coordination among leading AI companies, an initiative backed by OpenAI CEO Sam Altman and Elon Musk.
Meta CEO Mark Zuckerberg spoke out against a coordinated slowdown in the development of advanced artificial intelligence systems, arguing that each laboratory should decide how quickly it can train models safely. His remarks came amid a broader debate over whether developers should slow work on the most powerful AI models.
Zuckerberg said companies already have a natural incentive to improve product safety. Users will not choose agents that act against their intentions, he said, making trust and alignment with human goals competitive advantages.
Meta delayed Muse over safety concerns
Meta previously delayed the release of Muse by several months to focus on the AI agent’s safety and protection, Zuckerberg said. He stressed that Meta made the decision independently and did not ask other companies to pause their own work.
Meta opened access to Muse on September 9. The agent can work with email, calendars, browsers and connected services.
The company said Muse operates in a dedicated cloud virtual machine. A separate Sentinel agent monitors its actions on external platforms, and Meta offered rewards of up to $300,000 for identifying vulnerabilities in the system.
The Muse delay was not the only instance in which Meta’s AI plans fell short of its initial expectations. Zuckerberg previously acknowledged that the development of AI agents had not accelerated as quickly as company leaders had hoped.
Amid weaker-than-expected productivity growth, Meta also abandoned a second wave of large-scale workforce restructuring.
Industry debate over coordinated action
Anthropic CEO Dario Amodei previously proposed a three-step plan that includes independent audits and coordination among leading AI companies. OpenAI CEO Sam Altman and Elon Musk backed the initiative.
Zuckerberg’s position does not rule out slowing AI development when a company considers it necessary. Instead, he argues that individual companies should reduce their pace independently rather than through an industry-wide pause, citing Meta’s decision to delay Muse as an example.
Source: Incrypted
