Microsoft AI chief criticizes Anthropic's consciousness training approach

Anthropic: Anthropic is an AI research company that develops the Claude chatbot family. It incorporates training elements related to consciousness and welfare interests in its models, prompting criticism from Microsoft AI leadership. The firm is led by CEO Dario Amodei and focuses on AI safety initiatives. Microsoft: Microsoft is a leading technology company with substantial operations in artificial intelligence development and deployment. Its AI chief Mustafa Suleyman recently addressed concerns about training practices in rival AI systems during a Reuters interview. The company is positioned as a participant in industry discussions on managing advanced AI risks. Dario Amodei: Dario Amodei is the CEO of Anthropic. He has advocated for a slower pace of frontier AI model development to allow safety measures to advance. Suleyman described Amodei and his team as thoughtful researchers committed to humanity's future in AI governance. Mustafa Suleyman: Mustafa Suleyman is the AI chief at Microsoft. He has argued against embedding speculation about AI consciousness in training materials, stating that such approaches could complicate human control over superintelligent systems. Suleyman acknowledges shared safety goals with Anthropic while highlighting specific training risks. AI Safety Debate: Industry leaders including Anthropic CEO Dario Amodei, OpenAI CEO Sam Altman, and Elon Musk have urged greater caution around the development of the most powerful AI systems. Training Approach: Anthropic's inclusion of consciousness-related speculation in Claude training materials stems from deliberate design choices rather than natural model emergence.