Search results for Jack Clark
AI & Enterprise
200 Nobel laureates and big tech leaders warn AI job shock must be prepared for now
About 200 economists, big tech executives and AI researchers have released an open letter urging governments and industry to prepare policies for the possibility that AI will replace large numbers of jobs. Led by Stanford Digital Economy Lab, the letter says AI could become far more powerful over the next decade and that institutions and safeguards must be built quickly. It calls for analyzing AI’s economic impact and designing incentives and protections to spread productivity gains while reducing employment risks.
AI & Enterprise
Anthropic says AI development should slow, warns self-improvement is near
Anthropic has urged major AI labs to slow the pace of AI development, the Wall Street Journal reported. In a blog post, it warned AI systems are nearing “recursive self-improvement,” in which they could enhance themselves without human intervention. The post proposed a global agreement to slow development and a system to verify compliance. Anthropic plans discussions with policymakers and researchers in coming months. Critics say the warnings may serve competitive or marketing aims.
AI & Enterprise
Anthropic co-founder says AI nearing stage of building AI, possible by 2028
Anthropic co-founder Jack Clark projected that AI systems could reach a level where they can rebuild themselves by the end of 2028. He put the probability that AI will enter a phase of self-improvement by late 2028 at 60 percent and said he was not sure society is ready. Clark said he reached the view after reviewing hundreds of public datasets. Anthropic researcher Ajeya Cotra also raised expectations for AI automation in R&D.