Microsoft AI chief executive Mustafa Suleyman has warned that the development of increasingly autonomous artificial intelligence could eventually result in a new “silicon species” capable of competing with humans for resources.
Speaking to the BBC’s Today programme, Suleyman said concerns about the potential dangers of AI were justified, but stressed that practical safeguards could be put in place to keep the technology under human control.
He criticised what he described as efforts by rival AI company Anthropic to give its AI systems human-like characteristics, arguing that treating AI as if it were conscious or had its own desires could create serious risks.
Suleyman said technology companies should be cautious about developing systems that can set their own objectives, generate income and own assets.
He warned that giving AI such levels of autonomy could amount to “seeding a new silicon species” that could eventually compete with humans for resources.
His comments follow an essay he published earlier this week in which he questioned Anthropic’s approach to training its Claude AI models.
While praising Anthropic chief executive Dario Amodei and his team as “intellectually honest people”, Suleyman criticised the practice of anthropomorphising AI by presenting systems as having their own desires, values or sense of identity.
“AIs are not conscious,” he wrote, arguing that current systems do not experience feelings or have their own motivations.
Suleyman has called for greater transparency in the development of AI, including independent assessments of how systems behave and stronger tools to monitor and control them.
He also stressed the importance of AI alignment, a field focused on ensuring advanced AI systems operate according to human values and objectives.
Suleyman said AI systems should remain “subordinate” to humanity and argued that countries and technology companies should share an interest in ensuring AI remains safe and controllable.
The debate over how much independence should be given to AI systems is becoming increasingly prominent within the technology industry.
Microsoft established its own superintelligence team in 2025 as it continues to invest heavily in advanced AI development.
The company has also published a draft Humanist AI Code of Conduct, outlining its approach to developing advanced AI. The framework promotes the concept of “humanist superintelligence” — highly advanced AI designed to work for people, operate within defined limits and remain under human control.
University of Southampton computer science professor Dame Wendy Hall described Suleyman’s comments as an important international conversation, while cautioning against exaggerated claims about AI risks.
The debate highlights a central challenge for the technology industry: developing increasingly capable AI systems while ensuring they remain subject to human oversight and control.