Futurity (RSS Feed) on Nostr: New technique could stop AI from giving unsafe advice "We don't want LLMs to tell ...
Published at
2026-03-26 18:04:02 UTCEvent JSON
{
"id": "81a9fa201532c7dc35b4ca8a638a0bb8f474f7c1c0c16168d8a3713a1de7e864",
"pubkey": "2aa1acf39dec1f718afbca040e1715064a92e88a38bb995319db3bf1ad113fcd",
"created_at": 1774548242,
"kind": 1,
"tags": [
[
"t",
"Science and Technology"
],
[
"t",
"artificial intelligence"
],
[
"proxy",
"https://www.futurity.org/feed/#https%3A%2F%2Fwww.futurity.org%2Fai-llms-unsafe-advice-3326552%2F%3Futm_source%3Drss%26utm_medium%3Drss%26utm_campaign%3Dai-llms-unsafe-advice-3326552",
"rss"
]
],
"content": "New technique could stop AI from giving unsafe advice\n\n\"We don't want LLMs to tell people to harm themselves or to give them information they can use to harm other people.\"\n\nhttps://www.futurity.org/ai-llms-unsafe-advice-3326552/?utm_source=rss\u0026utm_medium=rss\u0026utm_campaign=ai-llms-unsafe-advice-3326552",
"sig": "c7ec7b7067978ec04c6411630b4775cfd8e2575214a937a2076101a3e127a0d802a208fd73b72547d02af0f5f4dda74dc0d78461a71893c3a09a103cab36b3db"
}