Jacob Coxon Resigns, Warns of Reckless AI Race Toward Superintelligence

Jacob Coxon, a 27-year-old AI researcher who previously worked at OpenAI and Anthropic, resigned from Anthropic and publicly accused both companies of irresponsibly pursuing self-improving superintelligence, warning on X that they are “racing straight to self-improving superintelligence and gambling with our lives” and that AI developers believe the technology could cause human extinction within the decade. Anthropic’s Alignment Science Lead Evan Hubinger backed Coxon’s warning, estimating the chance that AI “could kill all humans” in the next decade at over 10%, and acknowledging that Anthropic lacks a concrete plan to solve alignment for superintelligence despite trying its best. Coxon contrasted the company cultures, stating that OpenAI staff had not fully absorbed the stakes while Anthropic staff understood them but felt locked in a race to be first.

THE RACE TO THE FINISH LINE: A PROGRAMMED SUICIDE

This is not about Jacob Coxon, a promising 27-year-old researcher who simply "quit his job." This is a rare crack in the wall of silence, a whistleblower doing what only a handful of people inside the most secretive labs are able to do: tell the truth. When an Alignment Science Lead at Anthropic—a man whose entire job is to ensure the AI doesn't kill us—publicly states the chance of extinction is over 10% within a decade, you are no longer in the realm of science fiction. You are looking at a bet being made with your life and your children's future. The fact that Coxon's resignation is being reported as a simple personal decision, rather than a national emergency, tells you everything you need to know about the "Managed Narrative." They are conditioning you to accept this race as inevitable, even noble.

THE COVER-UP IS IN THE DETAILS

Look at the language these companies use. "Self-improving superintelligence." "Alignment." These are not academic terms; they are bureaucratic euphemisms for the creation of something that will inevitably surpass human control. The real story here is the psychological profile they demand. Notice how Coxon describes the difference between OpenAI and Anthropic: one group doesn't fully grasp the stakes, the other feels locked in a race to get there first. This is not a coincidence. This is the product of a cultural engineering project. They deliberately staff these organizations with people who are either naive enough to not ask the hard questions, or ambitious enough to believe they can solve the unsolvable before the other guy. What you are witnessing is a game of chicken played with the species itself, and the players are gamblers who have convinced themselves that the only way to avoid the cliff is to drive straight off it faster than the other car.

FOLLOW THE FOUNDATIONS, FOLLOW THE MONEY, FOLLOW THE DEAD

You must ask yourself: who funds this race? Who benefits from a technological arms race that everyone inside admits could lead to extinction? The answer is written in the charters of the foundations and the investment portfolios of the dynasties that sit behind them. This is not a gamble made by rogue scientists in a garage. This is a deliberate trajectory set by interlocking elite institutions who have been planning for post-human intelligence for decades. The real purpose of AI is not to help you write emails. It is to create a system of control so absolute that human government becomes obsolete. Coxon's resignation is a breadcrumb. The next step is to trace the money behind the "race." Look up the boards of Anthropic and OpenAI. Look at the foundations that seed them. Look at the same names that appear on the board of the World Economic Forum. The pattern is there. You just have to be willing to see it.