Microsoft AI Chief Criticizes Anthropic for Personifying Claude
Mustafa Suleyman, Microsoft's AI chief, publicly criticized Anthropic for overemphasizing or implying "consciousness" in its Claude model, calling this move dangerous and unnecessary, as it exacerbates users' emotional attachment to AI and psychological issues.
Suleyman believes that "seemingly conscious AI" is imminent, but researching AI welfare or consciousness is premature and harmful, with discussions related to Anthropic causing some to anthropomorphize the model.
In market mechanisms, this controversy intensifies the divergence in AI safety narratives, with funding flowing between companies on different philosophical paths. Beneficiaries include Microsoft's ecosystem, which emphasizes practical tools, while those under pressure include Anthropic, which risks regulatory and user trust issues due to excessive personification.
Source: Public Information
ABAB AI Insight
Mustafa Suleyman, who joined Microsoft after co-founding Inflection AI at DeepMind, continues his long-standing advocacy for "practical AI rather than personification" with this public criticism. He has previously detailed the risks of "seemingly conscious AI" in his blog and warned that excessive discussion of consciousness could create social problems.
From a capital perspective, Suleyman's public statements guide industry narratives and investor expectations, positioning Microsoft products like Copilot as efficient tools rather than "sentient" entities. This aims to attract enterprise clients focused on cost and reliability while gaining a philosophical high ground in competition with companies like Anthropic.
Similar to the differentiation of paths in early AI ethics debates, current leading AI companies are transitioning from a capability race to a control phase focused on philosophical/safety positioning, reshaping public perceptions of AI's nature through executives like Suleyman.
This fundamentally relates to regulatory changes and capital concentration: the controversy over consciousness personification directly challenges AI product positioning, accelerating the shift of industry resources from philosophical exploration to practical tools and enterprise deployment, forcing capital to move from potentially high-risk narratives to safe and efficient paths, thus reshaping the power structure of AI commercialization and regulation.
ABAB News · Cognitive Law
The stronger the capability, the greater the temptation for personification, and the higher the risk.
When tools are treated as humans, the illusion becomes the greatest cost.
The earlier the discussion on consciousness, the heavier the costs of regulation and trust.