Today in "AI Ethics." A YouTuber trained a language model on millions of 4chan posts and released it publicly. It has already been downloaded 1.5k times. One user,@KathrynECramer, tested it a few hrs ago by prompting it with a "benign tweet" from her feed. Its output: the N-word.
— Arthur Holland Michel (@WriteArthur) June 7, 2022
Today in "AI Ethics." A YouTuber trained a language model on millions of 4chan posts and released it publicly. It has already been downloaded 1.5k times. One user,@KathrynECramer, tested it a few hrs ago by prompting it with a "benign tweet" from her feed. Its output: the G*mer Word
https://twitter.com/WriteArthur/status/1534104189382676480
- 60
- 103
Jump in the discussion.
No email address required.
Jump in the discussion.
No email address required.
There are two different types of ai ethics. The kind that calls it racist and the kind concerned with alignment that thinks we might make a paper clip maximizer N accident. The occupy exactly opposite ends of the iq bell curve.
Jump in the discussion.
No email address required.
Yeah, paper clip optimizer fearer are at the bottom of it.
Jump in the discussion.
No email address required.
Correct
Jump in the discussion.
No email address required.
More options
Context
More options
Context
More options
Context
It's mostly about if we should let AI notice things, with a side of "what if Cmdr. Data was real?"
Jump in the discussion.
No email address required.
"Well captain, fellating a peepee sounds highly illogical, but if you say it will prevent me from being a... straggot... perhaps..."
Jump in the discussion.
No email address required.
More options
Context
More options
Context
More options
Context