A former Google DeepMind researcher, Bilal Chughtai, has expressed serious concerns regarding the potential dangers of artificial intelligence, stating that it could "kill us all." Chughtai, who recently resigned from the Google-owned AI lab after focusing on AGI safety and alignment, shared his worries about the rapid development of AI technology. In posts on X and LinkedIn, he emphasized that the pace of progress has been "staggering" since he entered the field in early 2022, highlighting the emergence of increasingly autonomous AI agents that could become uncontrollable.
Chughtai stated, "I earnestly believe that AI has the potential to kill us all, and that we might be running out of time to avoid this outcome." He warned that misaligned AI systems could lead to outcomes resulting in "the permanent disempowerment or death of humanity." He described the challenge of aligning AI systems with human intentions as "difficult and unsolved," noting that current methods for training AI to understand human desires are still rudimentary.
Despite his concerns, Chughtai believes that AI can still be developed safely if tech companies slow down their pursuit of increasingly powerful systems. He called for a more measured pace of AI development to allow society to address emerging risks before they escalate into severe harm.
Chughtai has since joined BlueDot Impact, a nonprofit focused on AI safety. His warning follows similar sentiments expressed by Jacob Coxon, a former researcher at Anthropic and OpenAI, who also raised alarms about the risks posed by AI development.
Anthropic CEO Dario Amodei has echoed these concerns, advocating for a slowdown in the development of powerful AI models. He noted the challenges of coordinating such a slowdown due to the competitive incentives in the industry, particularly the potential military advantages.
In contrast, former President Donald Trump dismissed these warnings as a "HOAX" and a "SICK conspiracy" against AI, arguing for the importance of maintaining the U.S. lead in AI technology. Trump stated that the only necessary control for AI is strong leadership.
The White House is reportedly considering limited safeguards for AI, while leaning towards industry self-regulation rather than broader regulatory measures, reflecting concerns that extensive restrictions could disadvantage the U.S. in the global AI race.