Zuckerberg's Open-Weights Manifesto Is Also Meta's Strategy

Per the Wall Street Journal's opinion pages, Mark Zuckerberg published an essay this week titled "The AI Future Is for Everyone," arguing that the central question of the moment is not whether superintelligence arrives but who gets access to it. Superintelligence restricted to a handful of institutions, he writes, is more dangerous than superintelligence distributed widely, and Meta will build toward three principles: empowerment, invention, and balance of power.

It is the second open-weights manifesto from a lab chief in a week, following Jensen Huang's "Open Weights and American AI Leadership," published as his first ever post on X four days earlier. The industry's most powerful figures are converging on openness as public doctrine, and the timing is not accidental. Washington is actively writing AI policy on export rules, procurement, and liability, and these essays are briefs filed in that proceeding as much as they are philosophy.

The strategic read deserves equal weight. Open weights commoditize the model moats of Meta's closed rivals while Meta monetizes distribution and attention, so the safety argument and the competitive strategy point in the same direction. The concentration critique is nonetheless real: AI capex circulates among a handful of interlocking firms, the leading labs' valuations assume the gains stay concentrated, and "balance of power" is a fair description of what open distribution would do to that arrangement. The claim that open is safer remains genuinely disputed among security researchers, and the honest test of any lab's openness doctrine is what it releases when releasing costs it something.

What to watch is whether policy adopts the framing. If Washington treats open weights as a strategic asset rather than a proliferation risk, the essays did their job.