Microsoft AI Chief Mustafa Suleyman warned that the global race to advance artificial intelligence should not come at the cost of safety measures for increasingly capable AI systems. In a September 20 interview with CNN, Suleyman said competition with China should not become a reason to delay safeguards.
Suleyman said US developers can continue advancing artificial intelligence while ensuring people remain in control of AI systems. He also supported efforts to establish safety agreements and said the industry could benefit from practices developed in China.
His comments come as Washington debates the role of federal oversight in advanced AI development. President Donald Trump has argued that the United States should maintain its lead over China in AI, while several technology leaders continue to call for additional safeguards.
Reuters reported that AI safety and competition with China remain connected to upcoming US-China discussions.
Microsoft’s approach to AI safety includes a proposed code of conduct that keeps AI systems responsive to human direction. The framework requires AI systems to accept correction and shutdown commands, reinforcing the company’s focus on human control.
Microsoft AI published the code on September 15, stating that people should remain more important than AI. The proposal focuses on safety and human control as priorities for the company’s AI models and remains open for public feedback.
Broader AI rules could also affect development and compliance costs across the technology industry. Alphabet and Meta Platforms could face similar considerations as they expand their AI products.
Also Read: Microsoft’s Hyderabad Cloud Region Expands AI Compute Capacity
Suleyman’s comments add to a wider debate over how governments and technology companies should balance rapid AI development with safeguards.
His position centers on continuing technological progress without sacrificing human oversight. The debate matters more as AI systems become more capable and companies explore how much autonomy their models should have.
Suleyman also raised concerns about training AI systems around concepts such as consciousness and welfare. He argued that such approaches could make future systems more difficult for humans to control or shut down.
The discussion highlights a central challenge for the AI industry - advancing powerful systems while ensuring humans retain meaningful control over their development and use.