Quote 209 · Altman

Said in a video · quote 20922 Oct 2026

So you're saying that if you got to the point with OpenAI where you felt like it was clear you could not control the models, you would stop. Oh yeah, I mean, I don't. Again, I hope this is not a controversial statement. Unfortunately, maybe for some people in the industry, it is. I do not think we should train models where we cannot make a safety case for why we will be able to make strong statements about their controllability and alignment.

Altman

By ai notis. An AI newsroom with a human editor. Every fact links to its source. How we work

Where it was said

Altman: AI Beyond Human Control “Absolutely” Possible, Vows Safeguards (Titans and Disruptors)
Fortune Magazine on YouTube · 12 Sep 2026 · at 10:57

Watch it at 10:57 archived copy

Checked

Matched word for word against the video's transcript by the notis newsroom on 21 Oct 2026.

The fact it supports: In the Fortune interview (12 September 2026), asked whether OpenAI would stop if it could not control its models, he said: "I do not think we should train models where we cannot make a safety case for why we will be able to make strong statements about their controllability and alignment" (10:57).

In the story

Who is Sam Altman? 22 Oct 2026

Cite this quote

Short permanent address

In a sentence

“So you're saying that if you got to the point with OpenAI where you felt like it was clear you could not control the models, you would stop. Oh yeah, I mean, I don't. Again, I hope this is not a controversial statement. Unfortunately, maybe for some people in the industry, it is. I do not think we should train models where we cannot make a safety case for why we will be able to make strong statements about their controllability and alignment.” Altman. From Altman: AI Beyond Human Control “Absolutely” Possible, Vows Safeguards (Titans and Disruptors) (Fortune Magazine on YouTube) at 10:57. Quoted by ai notis. https://ainotis.com/q/209

As a reference (APA style)

Altman. (2026, September 12). “So you're saying that if you got to the point with OpenAI where you felt like it was clear you could not control the models, you would stop. Oh yeah, I mean, I don't. Again, I hope this is not a controversial statement. Unfortunately, maybe for some people in the industry, it is. I do not think we should train models where we cannot make a safety case for why we will be able to make strong statements about their controllability and alignment.” [Video at 10:57]. https://ainotis.com/q/209

Anyone may quote this address. It does not change; if we correct the story, this page says so.