Andrew Ng, who helped starting up Google Mind and Coursera, knowledgeable Bloomberg TV that warnings of human extinction from AI are important extra science fiction than science, and that the industry stokes them for publicity and to shape regulation. The EU AI Act already lists every the engineering risks he needs prioritised and the loss-of-shield a watch on risks he dismisses, in the connected recital.
Andrew Ng has known as warnings of existential AI threat “important extra science fiction than science“, in an interview with Bloomberg TV. He helped starting up Google Mind and Coursera, and worked with every Dario Amodei and Sam Altman earlier in their careers.
He went additional than disagreeing.
Ng stated the industry stoked catastrophe fears earlier in the reveal to salvage publicity and shape regulation, and that the connected thing is going down now. The wave he blueprint started when an Anthropic researcher resigned on 8 September, accusing his worn employers of gambling with americans’s lives.
The accusation is no longer any longer new.
Ng knowledgeable a US Senate discussion board in December 2023 that fats companies hype apprehension to create rules that would possibly “pull up the ladder” towards new entrants. In the connected commentary he put the percentages of AI causing human extinction at roughly one in 10 million over a century.
Three years on, the wording has barely moved.
What has changed is the quantity. The resignation post reached 150 million views, and within days tech workers had turned into the structure into an in-silly myth, swapping the extinction line for absurd ones.
Ng reads that as a campaign. Our occupy reporting found imitation rather than coordination.
Europe settled the argument in regulations earlier than it broke out. The AI Act’s checklist of systemic risks covers diminished boundaries to chemical, biological, radiological and nuclear weapons, cyber ability, disinformation and most important accidents in indispensable sectors.
It furthermore covers loss of shield a watch on. The an identical recital names unintended considerations with shield a watch on touching on to alignment with human intent, and models making copies of themselves. Both the dangers Ng needs prioritised and the one he calls fiction sit in a single paragraph of European regulations.
The tasks are already live. The Commission’s tech chief Henna Virkkunen stated this month that EU regulations requires Anthropic and OpenAI to evaluate the threat of losing shield a watch on of a mannequin, and that right here is no longer any longer the case globally.
That’s Ng’s functional engineering, made compulsory.
These tasks fully apply above a compute threshold, so they plunge on the supreme builders and on no one else. That is the shape of the ladder Ng described in 2023, constructed by a regulator rather than by a lab.
