finnborge

10 ms·

1. 1.▲
   by finnborge 4mo ago

   (( I started writing a simple comment and it spun out into a larger rumination. The following is not discussed in the article )) What _content_ will bots/LLMs actually have access to in... 3 years? PubMed, Wikipedia, and various comp
2. 2.▲
   by finnborge 4mo ago

   I understand why the conversation on this article looks like it does, but the study is specifically focused on the potential for LLMs to operate as tutors for law students. I enjoy the extrapolation out to whether LLMs will replace lawyers
3. 3.▲
   by finnborge 11mo ago

   It already is "just" information retrieval, just with stochastic threads refining the geometry of the information.
4. 4.▲
   by finnborge 11mo ago

   I think this is well illustrated in a lot of science fiction. Irregular or abstract tasks are fairly efficiently articulated in speech, just like the ones you provided. Simpler, repetitive ones are not. Imagine having to ask your shower
5. 5.▲
   by finnborge 11mo ago

   My most charitable interpretation of the perceived misunderstanding is that the intent was to frame developers as "the user." This project would be the developer tool used to produce interactive tools for end users. More practical
6. 6.▲
   by finnborge 11mo ago

   At this extreme, I think we'd end up relying on backup snapshots. Faulty outcomes are not debugged. They, and the ecosystem that produced them, are just erased. The ecosystem is then returned to its previous state. Kind of like savi
7. 7.▲
   by finnborge 11mo ago

   In N years the idea of requiring a rigid API contract between systems may be as ridiculous as a Panda being unable to understand that Bamboo is food unless it is planted in the ground. Abstractly, who cares what format the information is sh
8. 8.▲
   by finnborge 11mo ago

   If you haven't already seen the DeepSeek OCR paper [1], images can be profoundly more token-efficient encodings of information than even CSVs! [1]: https://github.com/deepseek-ai/DeepSeek-OCR/blob/main&#x
9. 9.▲
   by finnborge 11mo ago

   I'm not sure I follow this entirely, but if the assertion is that "everything is math" then yeah, I totally agree. Where I think language operates here is as the medium best situated to assign objects to locations in vector
10. 10.▲
    by finnborge 11mo ago

    This is amazing. It very creatively emphasizes how our definition of "boilerplate code" will shift over time. Another layer of abstraction would be running N of these, sandboxed, responding to each request, and then serving whic