Bilal Chughtai, a former research engineer at Google DeepMind, has publicly warned that increasingly advanced artificial intelligence could pose an extreme threat to humanity. Chughtai previously worked on AGI safety and alignment research before leaving DeepMind in July 2026.
His comments come amid a growing debate within the AI industry over how quickly frontier systems are advancing and whether safety research is keeping pace. Chughtai said he believes AI has the potential to cause catastrophic harm and that time may be running short to prevent such an outcome.
The warning is his personal assessment, rather than a prediction that such an outcome will necessarily happen.
Chughtai’s warning follows comments from Anthropic researcher Jacob Coxon, who recently resigned and said some people developing advanced AI genuinely believe the technology could eventually kill everyone.
Anthropic scientist Evan Hubinger subsequently said he agreed with Coxon and estimated a greater than 10% chance of such an outcome within the next decade.
The discussion centres on the possibility of increasingly autonomous AI systems becoming capable of improving themselves or operating beyond effective human oversight.
Anthropic CEO Dario Amodei has called for a slower pace of advanced AI development, arguing that society needs time to address emerging risks.
His position has received public support from OpenAI CEO Sam Altman and Elon Musk, while other voices have questioned whether slowing development is practical or desirable.
Chughtai has argued that safer AI development remains possible, but believes companies need greater coordination rather than an unchecked race to build increasingly powerful systems. He has called for development to move at a pace that allows emerging risks to be addressed before serious harm occurs.
Bilal Chughtai, a former research engineer at Google DeepMind, has publicly warned that increasingly advanced artificial intelligence could pose an extreme threat to humanity.
Bilal Chughtai, a former research engineer at Google DeepMind, has publicly warned that increasingly advanced artificial intelligence could pose an extreme threat to humanity. Chughtai previously worked on AGI safety and alignment research before leaving DeepMind in July 2026.
His comments come amid a growing debate within the AI industry over how quickly frontier systems are advancing and whether safety research is keeping pace. Chughtai said he believes AI has the potential to cause catastrophic harm and that time may be running short to prevent such an outcome.
The warning is his personal assessment, rather than a prediction that such an outcome will necessarily happen.
Chughtai’s warning follows comments from Anthropic researcher Jacob Coxon, who recently resigned and said some people developing advanced AI genuinely believe the technology could eventually kill everyone.
Anthropic scientist Evan Hubinger subsequently said he agreed with Coxon and estimated a greater than 10% chance of such an outcome within the next decade.
The discussion centres on the possibility of increasingly autonomous AI systems becoming capable of improving themselves or operating beyond effective human oversight.
Anthropic CEO Dario Amodei has called for a slower pace of advanced AI development, arguing that society needs time to address emerging risks.
His position has received public support from OpenAI CEO Sam Altman and Elon Musk, while other voices have questioned whether slowing development is practical or desirable.
Chughtai has argued that safer AI development remains possible, but believes companies need greater coordination rather than an unchecked race to build increasingly powerful systems. He has called for development to move at a pace that allows emerging risks to be addressed before serious harm occurs.
