Anthropic Delays New AI Model Training to Bolster Security Before IPO

The maker of the Claude chatbot disclosed in the risk factors section of a regulatory filing that it has postponed pre-training a new frontier base model. The delay is intended to strengthen protections against the theft of model weights and intellectual property. The company's filing frames the decision as a security measure rather than a product setback. The disclosure comes as the company prepares for a public offering. Further details on the timing of resumed training were not provided in the excerpt.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.


Discussion (0)
Log in to join the discussion and vote.
Log in