NewsNTech
The gap between frontier AI capability and the safety infrastructure needed to responsibly deploy it has no agreed metric, no regulator, and until Saturday, no public consensus from the industry building it.
In nine hours, the chief executives of Anthropic, OpenAI, SpaceXAI, and Google DeepMind aligned behind a slower development pace, a coordination that would have been implausible given their years of competition and mutual hostility.
Anthropic CEO Dario Amodei opened the sequence at 10 a.m. ET with "We Must Pace the Frontier," a 3,800-word essay warning that without deliberate pacing, doom scenarios become probable.
Among them, he wrote, is a rogue-agent takeover of the entire internet, which he said could arrive as soon as six months from now.
Keep reading