UN AI experts warn against ‘apocalyptic’ rhetoric on risks
New York, USA—Several experts on a United Nations artificial intelligence (AI) committee warned on Monday against the apocalyptic rhetoric used by some professionals in the sector regarding AI risks.
“As scientists, it’s very important to invest in the science, to really understand what we know [from] what we don’t know, and investing in fear is not that helpful,” said Joelle Barral, an executive at Google DeepMind and a member of the International Scientific Panel on AI, an independent body.
She was responding to a question from Agence France-Presse (AFP) about recent public statements on AI’s potential dangers, particularly from Jacob Coxon.
The former Anthropic employee, who previously worked at OpenAI, said in early September that AI developers believed the technology “could kill us all by the end of the decade.”
A research engineer who left Google DeepMind in July made similar comments, saying it might already be “too late to avoid” humanity’s eradication by AI.
‘Truth more nuanced’
“We have to change the way we talk about AI risk because for a very long time, it’s either existential and everything is going to die and the killer robots are coming, or it doesn’t really matter,” Nobel Peace Prize laureate Maria Ressa, an investigative reporter, said on Monday at the same event, held in New York on the sidelines of the UN General Assembly.
“The truth though is far more nuanced and sophisticated,” she continued.
“You can’t turn that into ideology, which is, I think, where [the debate] has been heading.”
‘Rational discussion’
The International Scientific Panel on AI was created in 2025, with its members selected in February 2026.
Its mandate is to advance the scientific understanding of AI and inform UN deliberations.
Yoshua Bengio, considered one of the fathers of AI and also a cochair of the panel, said it was important to distinguish between what scientists actually know about AI risks and what can only be “plausibly extrapolated” from that evidence—an area where experts still disagree.
“The problem with shouting very loudly is that someone else can also shout very loudly,” he said.
What is needed instead is “a global rational discussion about what is going on, including about the uncertainty, including about the severity of the risks.”
