AI
“Sam Altman and Dario Amodei spoke at the U.N. Security Council, warning about AI risks; Dario Amodei, CEO of Anthropic, said that if managed poorly, AI could be a risk to humanity as a whole" (post caption also attributes to Altman: "We could lose control of the future to AI.")”
Plain restatementOn the stated occasion, OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei both addressed the UN Security Council about risks from increasingly capable AI. Amodei said that if AI is managed poorly it could be a risk to humanity as a whole. Altman said humans could lose control of the future to AI.
This checks out. On 23 September 2026 the UN Security Council held a high-level briefing on artificial intelligence and international security, convened by France, and the official UN record shows that OpenAI's Sam Altman and Anthropic's Dario Amodei both briefed the Council, along with researcher Yoshua Bengio and Hugging Face's Clément Delangue. Amodei said that if AI is managed poorly he believed it could be a risk to humanity as a whole, and Altman said we could lose control of the future to AI. Altman's line appears in the text OpenAI itself published of his remarks. Two small details the post leaves out: Altman appeared in person while Amodei spoke by video link, and Amodei phrased his warning as a personal belief. One framing point is worth noting: the post's headline says the executives asked the UN to control the technology, while what they actually asked for was common safety testing standards, an incident notification system and safeguards set by governments. The session produced no binding resolution, though the post does not claim otherwise.
Sam Altman and Dario Amodei [drifted from the evidence:] spoke at the U.N. Security Council, [drifted from the evidence:] warning about [drifted from the evidence:] AI risks; [drifted from the evidence:] Dario Amodei, [drifted from the evidence:] CEO of Anthropic, said that if managed poorly, [drifted from the evidence:] AI could be a risk to humanity as a whole" [drifted from the evidence:] (post caption also attributes to Altman: [drifted from the evidence:] "We could lose control of the future to AI.")
[added by the neutral restatement:] On the stated occasion, OpenAI CEO Sam Altman and [added by the neutral restatement:] Anthropic CEO Dario Amodei [added by the neutral restatement:] both addressed the UN Security Council about risks [added by the neutral restatement:] from increasingly capable AI. Amodei said that if [added by the neutral restatement:] AI is managed poorly [added by the neutral restatement:] it could be a risk to humanity as a whole. Altman [added by the neutral restatement:] said humans could lose control of the future to AI.
Red-tinted words in the claim drifted from the evidence. Green-tinted words are what a neutral restatement needs.
The trace / claim to source
- Altman and Amodei both briefed the UN Security Council at its 10228th meeting on 23 September 2026 on AI and international security, alongside Bengio and Delangue.
- Amodei said that if AI is managed poorly he believed it could be a risk to humanity as a whole. The post's paraphrase keeps both the condition and the scope intact.
- Altman said "We could lose control of the future to AI." The quotation matches the text OpenAI published of his remarks.
- Both used the appearance to ask governments to set international standards and safeguards rather than leaving the matter to companies alone.
- The image showing Altman in a Security Council setting is consistent with his in-person attendance that day.
- Omitted qualifier, minor: Amodei spoke by video link rather than in the chamber, which the post does not mention. "Spoke at the U.N. Security Council" remains a fair description of a remote briefing entered on the Council's record, and a reader's understanding of the claim does not change, so this is a missing detail rather than a distortion of meaning.
- Omitted qualifier, minor: Amodei's own sentence was hedged as a personal belief, "I even believe." The post's "saying AI could become a risk to humanity as a whole if it is managed poorly" drops the belief marker while keeping the conditional, which is a light compression rather than a strengthening.
- The image headline's characterization that the executives asked the UN to "control" the technology is the post's framing, not a quotation. What the evidence shows is a request for common testing standards, incident notification and government-set safeguards, which is narrower than a request for control of the technology. The caption itself does not repeat this framing.
- Whether the official UN transcript renders Amodei's sentence word for word as AP reported it. Every outlet reviewed traces the direct quote to AP wire copy, which is a single original quotation chain, though his briefing is separately documented as an official UN asset and the substance is corroborated by multiple independent reports of the session.
The meeting is on the official UN record. The Security Council held its 10228th meeting on "Maintenance of international peace and security: Artificial intelligence and international security," and the official transcript records a Council member thanking Yoshua Bengio, Sam Altman, Dario Amodei and Clément Delangue for their presentations. The Council's own monitor recorded in advance that France, holding the September presidency, was convening a high-level briefing on AI and international security during the high-level segment of the 81st General Assembly session, chaired by French Foreign Minister Jean-Noël Barrot. Both quotes in the post are real and closely matched. AP-sourced reporting carries Amodei's words as "If managed poorly, I even believe AI could be a risk to humanity as a whole," and Altman's as "We could lose control of the future to AI." Altman's line appears in OpenAI's own published text of his remarks, where he said there are two ways AI progress could go very badly, the first being that "we could lose control of the future to AI," with the risk that it moves so fast that people can no longer follow what is happening or intervene. The substance was a call for international coordination. Both CEOs told the Council the AI industry needs international standards and cooperation or global security is at risk, and they spoke separately to the Council while both urging world leaders to help manage the threats from rapid AI advances. Amodei called for common global standards for testing AI models for loss-of-control and misuse risks, and for a notification system for AI incidents significant to global security. Other briefers spoke in similar terms: Bengio told the Council that dangers from AI are "real and imminent."
Complete reasoning
The reply is formatted for pasting into the thread where the claim is circulating.
Compact share page: ai.trueseeker.com/s/64e9a2c5c04f/fL6-zata8b_ZfIuYqFDKsEPe_Oq
Ask this case
Answers come only from the case file above; nothing is added.
Did Sam Altman and Dario Amodei actually speak to the UN Security Council?
Yes. Both briefed the Council's 10228th meeting on 23 September 2026, which focused on artificial intelligence and international security, alongside researcher Yoshua Bengio and Hugging Face's Clément Delangue.
Are the quotes attributed to them accurate?
Yes. Amodei said that if AI is managed poorly he believed it could be a risk to humanity as a whole, and Altman said 'We could lose control of the future to AI,' which matches the text OpenAI published of his remarks.
Were both executives physically present in the chamber?
No. Altman appeared in person, while Amodei briefed the Council by video link. The post does not mention this difference, though calling it a UN Security Council briefing is still accurate.
Did they ask the UN to take control of AI technology?
Not exactly. The post's headline framing about asking the UN to 'control' the technology goes further than what they actually requested, which was common safety testing standards, an incident notification system, and safeguards set by governments.
Did this meeting result in any binding rules on AI?
The case file does not show that the session produced any binding resolution, though the post does not claim otherwise.