this post was submitted on 12 Feb 2026
56 points (96.7% liked)

Fuck AI

6301 readers
1253 users here now

"We did it, Patrick! We made a technological breakthrough!"

A place for all those who loathe AI to discuss things, post articles, and ridicule the AI hype. Proud supporter of working people. And proud booer of SXSW 2024.

AI, in this case, refers to LLMs, GPT technology, and anything listed as "AI" meant to increase market valuations.

founded 2 years ago
MODERATORS
 

Because they don't really search or index quality content (it's very expensive and hard to do) and their search implementation really sucks, they don't do any real improvement.

The process is like this:

  1. Take the user query and create 1-3 queries. For this process they use very stupid but fast and cheap models; because of that, sometimes they create very stupid search queries and, unlike a pro, they don't really know how to use search engines, like filtering, ranking, focusing...
  2. Combine these search results (it contains slop AI-generated summary pages, YouTube videos, maybe forums, maybe Wikipedia...).
  3. Use RAG with an LLM to find answers. LLMs will always try to find answers quickly, and instead of making a thinking loop in a long article they will use that slop page with a direct answer.

As you can see, there are many, many problems in this implementation:

  • The biggest problem is citation: they cite confidently but it's wrong.
  • They use low-quality data, like auto YouTube subtitles, improperly extracted tables and elements, content-farm sites, copycat sites, corporate blogs...
  • Their search results are low quality.
  • For the most important part (breaking down the user request) they use cheap, stupid models.
  • They handle all data in the same context instead of parallel requests (which is very expensive)

It's still strange to me: we always say "they have all the data, all the money, all the hardware..." but they still can't create a better AI search than random FOSS developers.

you are viewing a single comment's thread
view the rest of the comments
[โ€“] november@piefed.blahaj.zone 4 points 4 weeks ago (1 children)

Try searching for "seahorse emoji".

[โ€“] Zerush@lemmy.ml 0 points 3 weeks ago* (last edited 3 weeks ago)

Andi said:

The seahorse emoji phenomenon reveals a curious case of mass misremembrance - there has never been a seahorse emoji, yet both humans and AI language models firmly believe it exists[^1][^2].

When asked about the seahorse emoji, large language models respond with complete confidence that it exists, then spiral into confusion when trying to display it, often outputting random fish or horse emojis instead[^2]. This behavior stems from the models building an internal "seahorse + emoji" concept that crashes against reality when no matching token exists in their vocabulary[^1].

The technical explanation involves the models' logit lens - as they process the request through their layers, they construct a conceptual blend of "seahorse" and "emoji" that seems perfectly valid until the final output stage, where they're forced to select the closest available match[^1].

Many humans share this false memory, with Reddit threads and social media posts filled with people convinced they've seen a seahorse emoji before[^1]. While a seahorse emoji was proposed to the Unicode Consortium in 2018, it was rejected and has never actually existed[^2].

[^1]: Why do LLMs freak out over the seahorse emoji? - Theia Vogel [^2]: Emojipedia - Is There a Seahorse Emoji?

Adding that it don't exist in the official Unicode emoji pack, but they do in inofficial packs, eg. here