- Published on
Is Big Tech’s AI slowdown a safety pact or a cartel?
- Authors
- Name
- DeepHorus
Four of the most powerful figures in artificial intelligence have agreed to slow development, raising questions about whether the move is a genuine safety measure or an anti-competitive cartel. OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei, Google DeepMind cofounder Demis Hassabis, and SpaceX head Elon Musk reached a loose agreement over the weekend. Their stated goal is to "pace the frontier" of AI advancement.
The agreement, reported by The Verge on September 14, 2026, includes at least partial support for a proposal that would embed third-party auditors in AI labs. It also calls for regulating domestic labs and reaching a global slowdown agreement. The leaders have not yet committed to specific enforcement mechanisms.
Critics immediately spotted an ulterior motive. They argue that the slowdown could function as a cartel, allowing the biggest players to cement their dominance while blocking smaller competitors. By setting the pace collectively, the incumbents could avoid disruptive innovation from startups.
The proposal's backers frame it as a necessary safety measure. Embedding independent auditors would provide oversight of powerful models, while domestic regulation and international coordination would prevent a race to the bottom. The signatories claim that a coordinated pause is essential to manage existential risks.
However, the lack of "real teeth" is a central concern. Without binding commitments or verification mechanisms, the agreement may be little more than a public relations gesture. The source notes that the proposed solutions need real teeth to be credible.
The divide reflects a broader tension in AI policy between safety advocates and competition watchdogs. If the pact is genuine, it could set a precedent for global AI governance. If it is a cartel, it could trigger antitrust scrutiny from regulators in the US, EU, and beyond.
For now, the four leaders have only loosely agreed, leaving many details unresolved. The coming months will show whether they translate their rhetoric into enforceable rules. The outcome will shape the future of AI development and the balance of power in the industry.
