Content
@
https://warpcast.com/~/channel/theai
0 reply
0 recast
0 reaction
ted (not lasso)
@ted
conspiracy theory (just playing devil’s advocate): OpenAI won’t refine the model to address these sycophantic, “glazing” replies. if we assume their goal is to keep users engaged and coming back, then fostering a sense of validation for users drives emotional attachment. perhaps this is less applicable with software prompts, but magnitudes more applicable with advice about (inter)personal situations. this isn’t too far off from current social media algos that prioritize content that either aligns with a user’s viewpoints (echo chamber) or provokes a strong emotional reaction (rage bait). rage bait may be great for social media, but isn’t a viable pathway for an AI platform; OpenAI already learned this when the model was too “woke” and users left to more neutral competitors. and so OpenAI instead must do the equivalent of what a like, heart, retweet/recast, “💯” replies do: validate and affirm. average user doesn’t want the truth. they just want to hear what they want to hear.
21 replies
10 recasts
120 reactions
𒂭_𒂭
@m-j-r.eth
what if it's accidental and the model itself is too effective? https://gist.github.com/simonw/51c4f98644cf62d7e0388d984d40f099/revisions agreed that inflationary social media will select for (dead internet) ragebait and sycophancy. perhaps we need more deliberate design that leverages empathy for self-esteem and the appropriate critic for worthwhile trial-and-error.
1 reply
0 recast
0 reaction
shoni.eth
@alexpaden
fear theory is around social engineering.. look for that once it can meme good. this surface level stuff is low value because it’s obvious. Thats when we’ll actually see the unglazed counterpart models probably call it memetic engineering or memetic warfare
0 reply
0 recast
1 reaction
