Zuckerberg just proposed the AI industry police itself through "independent evaluators" — and he's quietly installing the dealmaker who brokered Trump's Middle East policy to make sure Washington lets him do it.
The Summary
- Meta's CEO called for AI labs to use independent evaluators and advisers to verify model safety, positioning himself as the responsible face of AI development
- Zuckerberg recruited Dina Powell McCormick to Meta's board — the Goldman Sachs executive who served as Trump's deputy national security adviser — during a walk at his Kauai compound last December
- The timing matters: OpenAI is in early talks for funding at a $1.2 trillion valuation, and Zuckerberg needs regulatory breathing room to compete
The Signal
Meta's public position on AI safety and its private strategy for navigating Washington are now perfectly aligned. Zuckerberg's call for independent AI safety evaluators sounds like responsible stewardship. His recruitment of Powell McCormick reveals what it actually is: a play to control the narrative before someone else does.
Powell McCormick isn't a tech executive. She's a geopolitical power broker. At Goldman Sachs, she ran sovereign wealth relationship strategy. In the Trump White House, she helped negotiate Middle East policy. Her addition to Meta's board isn't about understanding transformers or training runs. It's about making sure Meta's AI ambitions don't get strangled by regulation while competitors race ahead.
"The real safety debate isn't technical. It's political. And Zuckerberg just hired someone who knows how to win that fight."
The context makes this sharper. OpenAI's potential $1.2 trillion valuation puts pressure on every other lab to justify their own spend and trajectory. Google has DeepMind and decades of AI research. Anthropic has constitutional AI and the backing of Amazon. Meta has Llama, open weights, and now a direct line to the people who decide what "safe" means in Washington.
Zuckerberg's proposal for independent evaluators is smart precisely because it's vague. Who certifies the certifiers? Who decides what "safe" means? If you're Meta, you want those questions answered by frameworks you helped design, not frameworks imposed by competitors or regulators who don't understand the technology. Powell McCormick's role is to make sure Meta has a seat at that table before the table gets built.
Key strategic moves:
- Position Meta as the responsible AI company before regulation forces the issue
- Leverage Powell McCormick's relationships to shape policy frameworks
- Use "independent evaluation" language to forestall more restrictive government oversight
This isn't cynicism. It's how the game works. Every AI lab will claim their approach is the safe one. The difference is most of them will make that argument in blog posts and white papers. Meta will make it in rooms where actual policy gets written.
The Implication
Watch who else gets added to AI company boards in the next six months. The pattern Meta just set — hiring people with policy chops, not just technical credentials — will spread. If you're building in the agent economy, understand that your technical roadmap is now inseparable from your regulatory strategy. The companies that win won't just build the best models. They'll build the best relationships with the people who decide what "best" means.
For founders: if you're waiting for clear AI safety standards before you ship, you're waiting for a document that will be written by whoever moves first. Meta just moved.
Sources
Bloomberg Tech | Bloomberg: Meta CEO Zuckerberg Leans on Powell McCormick to Clear Path for AI Push