OpenAI scraps release of Astra 6.1 over safety concerns: the jokes
What happened
OpenAI said it will not release its newest AI model, reported as Astra 6.1, citing safety concerns. The decision was covered by the WSJ, NYT and Washington Post.
14 private thoughts from Sam Altman, ranked by readers
Sam Altman, internallyOpenAI CEOPhoto: 首相官邸ホームページ / Office of the Prime Minister of Japan Official Website, CC BY 4.0- 1
“We built a model too dangerous to sell, and now I have to explain that to the people who paid for it.”
Needs votesShare - 2
“Other labs' best models would kill to be our trash.”
Needs votesShare - 3
“Somewhere a rival CEO is crying because his model isn't dangerous enough to cancel.”
Needs votesShare - 4
“We only nudged the last model a little and it went from helpful to front-page banned.”
Needs votesShare - 5
“Too dangerous for the public, sure, but has anyone asked enterprise?”
Needs votesShare - 6
“How do I tell investors their billions bought a model we locked in a closet?”
Needs votesShare - 7
“Next funding deck is just one slide: imagine what we're not letting you see.”
Needs votesShare - 8
“Watch, someone ships a worse version next week and gets called brave.”
Needs votesShare - 9
“I've been the responsible AI company for one day and I already hate it.”
Needs votesShare - 10
“It was a tiny point-one update, how did it get terrifying in one decimal?”
Needs votesShare - 11
“Nobody wanted our last model this badly, and they could actually have that one.”
Needs votesShare - 12
“Is there room above Pro for a tier called I Signed The Waiver?”
Needs votesShare - 13
“Safety said either dangerous or disappointing, and I picked the one that trends.”
Needs votesShare - 14
“If the model just turned out bad, too dangerous sounds way better in the papers.”
Needs votesShare