←back to thread

Linda Yaccarino is leaving X

(www.nytimes.com)
542 points donohoe | 1 comments | 09 Jul 25 14:52 UTC | HN request time: 0.233s | source
Show context
steveBK123[dead post] ◴[09 Jul 25 16:10 UTC] No.44511769[source]▶
[flagged]
ceejayoz ◴[09 Jul 25 16:22 UTC] No.44511884[source]▶
The other LLMs don't have a "disbelieve reputable sources" unsafety prompt added at the owner's instructions.
replies(2): >>44511947 #>>44512590 #
steveBK123 ◴[09 Jul 25 16:27 UTC] No.44511947[source]▶
It's gotta be more than that too though. Maybe training data other companies won't touch? Hidden prompt they aren't publishing? Etc.

Clearly Musk has put his hand on the scale in multiple ways.

replies(4): >>44512280 #>>44512305 #>>44513674 #>>44515749 #
peab ◴[09 Jul 25 19:03 UTC] No.44513674[source]▶
I think it's more so that they push changes quickly without exhaustively testing. Compare that to Google, who sits on a model for years for fear of hurting their reputation, or OpenAI and Anthropic who extensively red teams models
replies(1): >>44515043 #
1. steveBK123 ◴[09 Jul 25 21:48 UTC] No.44515043[source]▶
Why does Grok keep "failing" in the same directional way if its just a testing issue?