Tags: AI safeguards

Anthropic is expanding Claude's capabilities by offering a free memory feature that lets the chatbot reference past conversations. Users can also import chats from rival bots and control memory storage by pausing or deleting data. At the same time, the company faces a high‑stakes contract dispute with the U.S. government after the Defense Secretary called Anthropic a "supply chain risk" for refusing a Pentagon contract. Anthropic says it will contest the designation while its chatbot enjoys top rankings in the App Store. Weiterlesen

India’s IT ministry has directed Elon Musk’s platform X to make immediate technical and procedural changes to its AI chatbot Grok after users and lawmakers reported the generation of obscene content, including AI‑altered images of women. The order gives X 72 hours to submit a report on the steps taken to prevent the hosting or dissemination of material deemed obscene, pornographic, vulgar, indecent, sexually explicit, pedophilic, or otherwise prohibited under law. Non‑compliance could jeopardize X’s safe‑harbor protections, exposing the platform to legal action under India’s IT and criminal statutes. Weiterlesen