Zuckerberg turns superintelligence into a debate about access, power, and control

Mark Zuckerberg has argued that the defining question around superintelligence is not when it will arrive, but who will have access to it.

In a Wall Street Journal essay, the Meta CEO makes the case for widely distributed AI rather than a small number of institutions controlling the next generation of capability. The essay is also a strategic intervention in one of the industry’s most important debates: who should shape access to advanced AI.

For enterprises, policymakers, and infrastructure providers, that question extends beyond models themselves to the compute, governance, and market structures that determine who can build, deploy, and benefit from them.

Zuckerberg Reframes the AI Debate Around Access

At the heart of Zuckerberg’s argument is a simple proposition.

“The defining question of our age isn’t whether superintelligence will exist, but who will have access to it,” he said in the essay.

Rather than focusing on existential risk or the capabilities of future systems, Zuckerberg has argued that the greater challenge is preventing advanced intelligence from becoming concentrated within a handful of organisations.

He presented the case for a philosophy built around individual empowerment, invention, and a balance of power, arguing that broader access to superintelligence would produce greater innovation and prosperity than centralised control.

The essay places access alongside safety as one of the central questions now shaping AI policy and industry strategy.

The Essay is Really About Power

Although it is framed around superintelligence, the essay is fundamentally about governance.

Zuckerberg argued that concentrated power can limit innovation and human potential, and that the same logic should apply to AI.

“The history of democracy and economics has shown that there is no single objective answer to how people define the best life, and therefore the best approach is letting people decide what matters to them,” said Zuckerberg.

The implication is that AI should not become an infrastructure controlled by a small number of providers or institutions.

Instead, Zuckerberg said that broader distribution of advanced capability would support competition, individual agency, and a lower concentration of influence over economic, scientific, or political outcomes.

Whether that model aligns with the commercial realities of frontier AI remains open to debate, but it places market structure alongside safety as a defining issue of the AI era.

Openness Becomes Part of the Security Debate

The essay also links AI access with cybersecurity.

Drawing on the history of open-source software, Zuckerberg argued that openness can strengthen defensive capability by allowing more organisations to inspect and improve critical systems.

“In most cases, like cybersecurity, the history of open-source software has shown that giving everyone full access to powerful systems will be the best way to protect safety and security over time.”

The argument echoed a view seen in parts of the cybersecurity community that transparency and shared tooling can improve resilience, particularly as AI agents become more deeply embedded in enterprise environments.

It also aligns with recent industry initiatives focused on open AI security infrastructure, where supporters argue that defenders need systems they can inspect, adapt, and operate within their own environments rather than relying solely on opaque external platforms.

Zuckerberg also acknowledged that some risks, including biological threats, will continue to require coordination between governments and industry over the deployment of increasingly capable models.

Automation or Empowerment?

The essay also challenges assumptions about AI’s impact on employment.

Rather than presenting automation as AI’s defining outcome, Zuckerberg said that superintelligence should expand human capability by helping people create new products, discover new knowledge and build businesses.

“If superintelligence is widely distributed, then I believe we will see more jobs in the future, not fewer,” stated Zuckerberg in the essay.

His argument is that reducing barriers to advanced AI could make entrepreneurship more accessible, enabling smaller organisations and individual creators to compete without requiring significant capital investment.

That alters the emphasis from automation alone to the question of who can use AI and how widely those benefits are shared.

Access Becomes the Next AI Battleground

Zuckerberg’s essay is best understood as a contribution to the emerging debate over AI power.

As AI capabilities advance, questions of access, governance and infrastructure are becoming as important as model performance.

For enterprises, the discussion centres on whether AI remains something organisations can deploy and govern on their own terms, or whether access becomes mediated through a small number of dominant platforms.

For policymakers, it raises a parallel question about market structure: how to encourage innovation while avoiding excessive concentration of technological power.

The debate over superintelligence is therefore no longer only about what AI will become. It is also about who controls the infrastructure beneath it, who benefits from its capabilities and how widely those benefits are shared.