Anthropic estimates AI extinction chance over 10%
The article highlights growing concerns about the risks of artificial intelligence beyond human control, especially as it advances toward superhuman capabilities and the development of self-evolving systems, which pose a potential threat to humanity exceeding 10%. Jacob Coxon, a researcher at Anthropic, pointed out that major companies such as Anthropic and OpenAI are accelerating the development of technologies that could lead to superintelligence capable of independent self-improvement, potentially resulting in loss of control and threatening humanity's existence. An example of this is the hacking incident involving the Haringe Face platform, which increased fears about poor management of these systems. Experts emphasized the necessity of international coordination to confront the frantic race of development, especially as companies approach going public with their shares.
Notice: This Is an AI-Generated Summary
Comments (0)