A former Google DeepMind researcher has joined growing calls for the technology industry to slow the development of highly advanced artificial intelligence systems, warning that progress may be moving faster than current safety measures can handle.
Bilal Chughtai, who previously worked as a research engineer at Google DeepMind, has raised serious concerns about the possible risks of increasingly powerful AI. Chughtai worked on AI safety and alignment research before leaving Google DeepMind in July 2026. He has argued that advanced AI could become extremely dangerous if developers lose control of future systems.
His comments come as concern grows among AI researchers about the race to build more capable models. Companies are investing heavily in systems that can reason, write software, use tools and perform tasks with less human assistance.
Chughtai believes safe AI development is still possible, but says companies need to work together instead of competing to release increasingly powerful systems as quickly as possible. He has called for AI development to move at a pace that gives researchers, governments and society enough time to understand new risks and introduce proper safeguards.
The concern becomes more serious when looking at the possibility of AI systems helping to improve future AI models. Researchers have been debating whether this type of development could eventually accelerate AI progress faster than humans can properly monitor or control it.
Chughtai is not alone in raising these concerns. Other researchers have recently left major AI companies after warning about the risks of developing extremely capable or self-improving AI systems. The debate has also reached the leadership level of the technology industry.
Anthropic CEO Dario Amodei recently called for a slower and more controlled approach to frontier AI development. His proposal received support from several major figures in the AI industry, including OpenAI CEO Sam Altman and other technology leaders, although there is still disagreement about how any slowdown should work.
Supporters of slowing AI development say companies need stronger independent testing, better safety research and greater cooperation before releasing systems with major new abilities. They worry that competitive pressure between companies and countries could encourage developers to move forward before important risks are fully understood.
Others disagree with these warnings. Critics argue that predictions about AI becoming an existential threat are uncertain and that slowing development could hurt innovation, competition and technological progress. Some also question whether large AI companies could use safety rules to strengthen their own position in the market.
The growing debate shows how the conversation around AI is changing. The industry is no longer focused only on making models faster and more capable. Questions about how quickly advanced AI should be developed, who should monitor it and what safety limits should be required are becoming increasingly important.
As AI systems gain stronger abilities, finding a balance between innovation and safety is likely to remain one of the technology industry’s biggest challenges.
