This coverage of COT in ML is misleadingly anthropomorphic. Have we really lost track of how these things work? Just because we call something "chain of thought" that doesn't make it ACTUAL chain of thought. Anthropic has always done this. #MLsec
And this is a usually excellent reporter falling prey to the nomenclature.
https://arstechnica.com/ai/2025/04/researchers-concerned-to-find-ai-models-hiding-their-true-reasoning-processes/?utm_brand=arstechnica&utm_social-type=owned&utm_source=mastodon&utm_medium=social
Gary McGraw
npub1as…z5ale
2025-04-10 22:42:44 UTC
Author Public Key
npub1as04fahu6mge8kxmx06r40eqg9j3c3ex5pz5jcccdlsedw9l7cssez5aleSeen on
wss://relay.momostr.pinkPublished at
2025-04-10 22:42:44 UTCEvent JSON
{
"id": "957a101b28e641bb690b114ba0fbde6c8997e400c0920b6b4d5d2f770dc092ca",
"pubkey": "ec1f54f6fcd6d193d8db33f43abf2041651c4726a0454963186fe196b8bff621",
"created_at": 1744324964,
"kind": 1,
"tags": [
[
"proxy",
"https://sigmoid.social/@cigitalgem/114316080842936397",
"web"
],
[
"t",
"mlsec"
],
[
"proxy",
"https://sigmoid.social/users/cigitalgem/statuses/114316080842936397",
"activitypub"
],
[
"L",
"pink.momostr"
],
[
"l",
"pink.momostr.activitypub:https://sigmoid.social/users/cigitalgem/statuses/114316080842936397",
"pink.momostr"
],
[
"-"
]
],
"content": "This coverage of COT in ML is misleadingly anthropomorphic. Have we really lost track of how these things work? Just because we call something \"chain of thought\" that doesn't make it ACTUAL chain of thought. Anthropic has always done this. #MLsec\n\nAnd this is a usually excellent reporter falling prey to the nomenclature. \n\nhttps://arstechnica.com/ai/2025/04/researchers-concerned-to-find-ai-models-hiding-their-true-reasoning-processes/?utm_brand=arstechnica\u0026utm_social-type=owned\u0026utm_source=mastodon\u0026utm_medium=social",
"sig": "f69ea37ce7b9b0892bbbcad638210225fe6f55eba2eaac2898d89f3cc7beeabd33828e53cec540f4d97abeb564b532e5b983909f608fefb5c31f2acc68e2f1ef"
}