WASHINGTON, DC - JANUARY 31: Mark Zuckerberg, CEO of Meta testifies before the Senate Judiciary Committee at the Dirksen Senate Office Building on January 31, 2024 in Washington, DC. The committee heard testimony from the heads of the largest tech firms on the dangers of child sexual exploitation on social media. (Photo by Alex Wong/Getty Images)
Getty Images
On July 28, Mark Zuckerberg published an op-ed in the Wall Street Journal arguing that superintelligence should be distributed as widely as possible and that concentrating it in a handful of institutions is the greatest risk AI poses to humanity.
On the same day, more than 1,000 employees at frontier AI labs including OpenAI, Anthropic and Google DeepMind published an open letter asking the U.S. government to build the tools needed to deliberately slow down AI development.
Was the discursive thread accidental, or emblematic of the two poles the AI industry is now bifurcated between?
What Zuckerberg Contended
“The defining question of our age isn’t whether superintelligence will exist, but who will have access to it,” Zuckerberg wrote in the Wall Street Journal.
“Will it be centralized and restricted to a few institutions, or will it be a tool that empowers everyone?"
Unsurprisingly, Zuckerberg answers his own question fusing libertarian and populist rhetoric. “I propose a philosophy based on individual empowerment as the source of prosperity, invention as the primary purpose of superintelligence, and balance of power as the foundation of safety,” he wrote.
MORE FOR YOU
“If only a handful of institutions have superintelligence, they will inevitably exercise a controlling influence over economics, science and politics.”
He also subverts some of the most prominent employment worries around AI. “I don’t understand why anyone who believes that AI will eliminate most jobs and much of humanity's relevance would rush to build that future,” he wrote. His counter is that the history of transformative technology — electricity, personal computing, flight — shows each wave producing more widely shared prosperity.
“Each time there is fear that people will be left behind. But each time humanity has come out with more people sharing greater prosperity, health and freedom. We believe this will be true with AI as well, and the abundance of the future can be shared by everyone.”
He also makes a more specific predictive assertion about jobs. “If superintelligence is widely distributed, then I believe we will see more jobs in the future, not fewer.”
Same-Day Counterpoint
On the same day, Pacing the Frontier — a public letter signed by more than 1,000 employees at frontier AI companies — went live.
“We request that the U.S. government support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development,” the letter stated.
The signatories are asking Washington to help build the steering wheel before the engine hits recursive gear — to make the option to slow down viable so that no single lab or country has to unilaterally sacrifice competitive ground to exercise it.
Within hours, OpenAI and Anthropic endorsed it as companies, too. Anthropic's post, reposted by co-founder Jack Clark, stated the petition was signed by its CEO, several co-founders and senior staff, and tied the ask to Anthropic's own recursive self-improvement research. Signatories include Dario Amodei, CEO of Anthropic; Jakub Pachocki, Chief Scientist at OpenAI; Mark Chen, Chief Research Officer at OpenAI; Shengjia Zhao, Chief Scientist at Meta AI; and Anca Dragan, Vice President of AI Safety and Alignment at Google.
“There is a real risk that capability development rapidly accelerates beyond our ability to understand or control the resulting systems,” the letter states.
While both pieces went live on the exact same day, they represent two philosophies that appear somewhat at odds over AI safety, regulation and its broader trajectory.
Competitive Subtext
Notably, Meta’s AI business strategy hinges on open-weight models — releasing model weights publicly so anyone can modify and run them. This strategy is different from that of closed-model labs like OpenAI and Anthropic whose commercial position hinges on maintaining proprietary access to their most capable systems.
“It’s no surprise that Zuckerberg believes Meta is the appropriate surface for, as he puts it, ‘delivering personal superintelligence to everyone,’” media analyst John Battelle noted. “The creator of the world’s most successful walled garden and owner of Instagram argues without a trace of irony that ‘humanity isn’t a monoculture’.”
The distribute-and-accelerate argument is also one of Meta’s strongest competitive positions — and calling for open access to superintelligence may not be a neutral philosophical stance when it is coming from a company committed to making its models openly available. In other words, while these notions sound highly philosophical and meta, it also heavily favors Meta’s current business strategy.
“The debate raging throughout Silicon Valley is being painted as geopolitical (we can’t let China win!), as existential (unfettered superintelligence will destroy humanity!), and as economic (data centers and billionaires are taking our jobs and stealing our future!) - all of these issues are worthy of debate and action,” noted Battelle. “But the recent bloviation from tech CEOs really comes down to one simple question: In a future dominated by AI, who will own the customer?”
Aligned With Everyone At Once? Impossible!
Zuckerberg’s vision is one where personal superintelligence becomes a consumer product, as available and varied as smartphones. The Pacing the Frontier signatories may not be opposing that vision all at once, but they do seem to be arguing that before anything else happens, governments need the ability and framework to hit the brakes if automated AI research starts compounding faster than human oversight can reasonably track.
“I think it is literally impossible to have a single benevolent superintelligence that is simultaneously aligned with everyone at once,” Zuckerberg said in an interview with The New York Times on the same day. His rivals, it appears, are more worried about the alternative.