Search AI Products and News

AI company Anthropic recently updated its "responsible scaling" policy, outlining which models require additional safety protections. This move aims to mitigate potential risks before new technologies are released. According to Anthropic's blog, if stress testing reveals an AI model could assist a "resource-constrained state actor" in developing chemical and biological weapons, Anthropic will delay the technology's public launch.

In the field of artificial intelligence, the startup Anthropic has recently made an exciting announcement. According to a latest report from CNBC, the company is expected to reach $1 billion in revenue this year, with an impressive annual growth rate of 1000%. This massive increase has shocked the entire industry, indicating strong demand for AI technology. Notably, 60% to 75% of Anthropic's revenue comes from the use of third-party application programming interfaces (APIs). This demonstrates

The competition in large models is fierce, with Llama-1 and 2k already outdated. The new generation of GPT-4 and Claude2.1 is leading the trend. The author discovered through extreme testing that most people did not use large models correctly and failed to unleash the AI's potential. The Claude team offers a simple solution: by adding one sentence, the performance jumped from 27% to 98%. Domestic large model companies also participated in the testing, with the Kimi team from Moon's Dark Side achieving good results using different methods.