On July 13, Stanford’s Digital Economy Lab published one of the shortest documents in the history of AI policy. Eighty-eight words. Four sentences. A title with no subtlety in it at all: We Must Act Now: A Statement on AI’s Transformation of the Economy.

Within days it had been signed by more than 200 economists and AI researchers, including sixteen Nobel laureates. The signatory count has since climbed toward two thousand.

The statement itself says roughly what you would expect. AI may become radically more powerful over the next decade. That could drive an economic transformation larger than the Industrial Revolution, but unfolding over a vastly shorter time frame. It warns of large-scale job displacement alongside the possibility of major gains in living standards. It calls for building incentives, guardrails, and institutions now rather than later.

If you have been reading about AI for the last three years, none of that is new. Warnings are cheap. There have been dozens.

The interesting part is not what the statement says. It is who put their name on it.

The two signatures that matter most

Among the signatories are Daron Acemoglu and Simon Johnson, the MIT economists who shared the 2024 Nobel Prize.

For years, Acemoglu was among the most prominent and rigorous skeptics of AI displacement predictions. He argued the productivity effects were being wildly oversold, that the economic modeling behind the doom forecasts was weak, that history suggested a much slower and messier transition than the technologists claimed. He was not a fringe voice. He was the serious counterweight the debate needed, and he made the strongest available case that everyone was getting carried away.

He signed.

This is what makes the statement worth paying attention to when the previous hundred warnings were not. A warning signed by people who always believed the warning tells you very little. A warning signed by the person who spent a decade arguing against it tells you something changed.

We tend to treat consistency as a virtue in public intellectuals, and there is something to that. But in questions of forecasting, the opposite is true. The person who never updates is not being principled; they are being uninformative. The valuable signal is when a careful skeptic looks at new evidence and says, publicly, that they were wrong about something. That costs them. That is why it means something.

What actually changed

It is worth being precise about why the profession moved, because “economists got scared” is not an explanation.

The organizers — Erik Brynjolfsson at Stanford, Ajay Agrawal at Toronto, Anton Korinek at Virginia, and Tom Cunningham at METR — framed the problem as a gap. Capabilities are now moving faster than economic understanding of what those capabilities do to labour markets. Brynjolfsson told the New York Times there had been “a notable change in the profession.”

The change is not that economists suddenly believe in science fiction. It is narrower and more interesting than that. The old skeptical position rested on a reasonable historical observation: technology has always displaced some work while creating more of it elsewhere, and the adjustment, though painful, has always been absorbable. Looms, tractors, spreadsheets. Every time, the doom forecasts were wrong.

The new position does not reject that history. It questions whether the timeline holds. The Industrial Revolution took roughly a century to reshape work. Institutions — unions, public education, labour law, welfare systems had decades to form around each shock. If a comparable transformation compresses into ten or fifteen years, the historical reassurance stops applying, not because the pattern is different but because there is no time for the absorbing mechanisms to build.

That is the actual argument. Not “AI is magic.” Rather: our institutions are calibrated to a speed of change that may no longer be the speed we are moving at.

The fair criticism

The statement has a real weakness, and it is worth naming rather than glossing over.

It does not propose anything specific. Critics have pointed out that it contains no policy recommendations, no proposed legislation, no mechanism. It calls for “incentives, guardrails and institutions” without specifying any. At least one prominent figure in AI declined to sign, and the objection that it amounts to a very decorated shrug is not unreasonable.

The defence is that consensus-building is itself the function. Getting two thousand economists, including former skeptics and executives from Google, Anthropic and OpenAI, to agree on a single framing is not nothing it establishes that preparation is overdue as a mainstream position rather than an activist one. Nobel laureate Michael Spence described it as a call for an “all hands on deck” approach.

Both readings are defensible. You can hold that the statement is an important signal and also that it is frustratingly empty of instruction. Those are not in conflict.

What this means if you are not an economist

Strip away the institutional drama and there is something practical here.

The statement’s core claim is about speed, not certainty. Nobody in that signatory list knows what happens next. What they are collectively saying is that the range of plausible outcomes has widened, the fast end of that range has become much more plausible than it was three years ago, and waiting for certainty before preparing is now a bad strategy.

That logic scales down to individuals surprisingly well. You do not need to believe AI will transform your industry within five years. You need to notice that the probability of it happening fast has gone up enough that the cost of being unprepared has gone up with it. The correct response to widening uncertainty is not panic and not dismissal. It is building optionality skills that transfer, work that compounds, a financial position that survives a disrupted year.

That is unglamorous advice. It is also what the entire statement amounts to, translated from the institutional register into a personal one.

The most decorated economists on earth just published four sentences saying they do not know how fast this goes, but they no longer feel comfortable assuming it goes slowly. That is not a prediction. It is a change in posture.

When the people who spent a decade telling you not to worry start telling you to prepare, the useful response isn’t fear. It’s attention.