Three people with every commercial reason to accelerate spent last week asking for a brake. Amodei, Altman and Musk converged publicly on slowing something down. Amodei’s post pulled 69 million views. Obama moved AI safeguards to the centre of Democratic policy.
Then Trump answered, per the FT: “We’re leading China in AI, and, frankly, I want to keep it that way, because whoever wins AI, wins.”
That settles the pause debate for this administration. It settles nothing for you.
Whether the warnings are right is a question you cannot answer, cannot influence, and cannot plan on. Every hour spent on it is an hour someone with a position wanted you to spend. The question you can answer, today, alone, in ninety minutes: which of your current commitments only make sense in one future?
The four futures
Most people model two: fine, or over. The probability lives in the middle.
1. Plateau. Capability hits a wall. What exists gets absorbed into ordinary software over a decade, like the cloud. Nobody writes essays about it.
2. Grinding diffusion. Capability keeps improving and deployment stays the bottleneck. Value moves to distribution and integration. China’s pivot from model racing to agent deployment points here. So does Meta’s claim that an agent swarm outperformed 100 engineers: the reported gain came from evaluation loops, meaning systems engineering rather than a smarter base model.
3. Fast, managed. Capability jumps and institutions keep pace. Microsoft’s public consultation on rules for its own superintelligence models is either the early shape of this or theatre.
4. Fast, unmanaged. Capability jumps and coordination fails. Trump’s rejection is this logic winning: if the US slows and China accelerates, a pause is a transfer.
My split, dated today: 15 / 55 / 20 / 10. Write down yours, because the number you’d defend publicly and the number you believe are different, and only the written one is useful in six months.
Every decision you make this year is a bet on one of these four. Most people have never named which.
Below the paywall
-
The monthly dashboard. Twelve indicators, where to find each, which scenario each confirms. Twenty minutes a month, and the anxiety becomes observation
-
The robust list. Six decisions that pay off in all four futures. Shorter and stranger than you’d expect
-
The fragile list. Commitments that assume one future without saying so, including three that look conservative and are the opposite
-
The career audit. Score your actual tasks on automation exposure and consequence, find the danger quadrant, move one task a quarter
-
The capital section. The market just had its best quarter in decades on the back of the thing being warned about. AI is now 25 to 33% of US GDP growth. What that does to your positioning, and the circularity number that breaks first
-
Carrying it. For readers who arrived from the anxiety rather than the strategy
Start your free 7-day trial
Cancel anytime. First subscribers get 50% off forever.
Read more
