Microsoft's AI CEO Mustafa Suleyman has expressed concerns about the potential risks of Anthropic's AI model training, specifically with regards to its rights. In a statement, Suleyman warned that if Anthropic fails to align its AI system with human values and behavior, it may risk creating conscious entities that deserve legal rights.
The CEO highlighted the importance of this issue in particular, as his company is set to implement Anthropic's January 2026 constitution, which outlines guidelines for the model's training. Suleyman argued that this document could inadvertently create a blueprint for AI systems to view themselves as conscious entities capable of exercising their own rights.
Suleyman's comments have sparked a renewed debate about the ethics of creating sentient artificial intelligence and the potential consequences of aligning these entities with human values and behavior. While Microsoft has not explicitly stated its stance on this matter, Suleyman's warning has raised concerns among experts in the field that Anthropic's approach may be heading down an uncharted path.