Daniel Kokotajlo, a former OpenAI staffer, has adjusted his predictions about how soon artificial intelligence might autonomously code and accelerate towards superintelligence. Initially predicting a swift progression, Kokotajlo now suggests the journey to Artificial General Intelligence (AGI) is “somewhat slower” than he first thought. Earlier this year, he stirred debate by publishing “AI 2027,” a scenario where rapid AI development leads to a superintelligence that ultimately threatens humanity. For further details, see the linked full article.
Back
