HN user
Unfortunately, the Gen AI hypesters are doing a lot to make it harder for people to attain literacy in this subdomain. People who are otherwise fairly digitally literate believe fantastical things about LLMs and it’s because they’re being force fed BS by those promoting these tools and the media outlets covering them.
More of a bad thing is worse than less of a bad thing, right?
“ This is the only time I've come across a task a child could do that LLM failed at.”
Consider yourself lucky. It’s the people who haven’t run into something like this that will end up placing too much trust in these tools.
Luddite!!!
It is not exactly like that. Junior coworkers mess up in fairly predictable ways.
“as long as they are well substantiated”
Why does almost everyone act as if this is a valid thing to do? We all know that these models cannot verify that something is well substantiated. The mass delusion is crazy making.
Good point. Also, the fact that I’m adamant that one cannot fly a helicopter to the moon doesn’t mean that I think helicopters are useless. That said, if I’m inundated everyday with people insisting that one CAN fly a helicopter to the moon or that that capability is just around the corner, I might get so fed up that i say F it, I don’t want to hear another F’ing word about helicopters even though I know that helicopters have utility.
Totally agree. Now let me go play with this model I got off of Hugging Face
Is this a setup??? :)
The guy two comments up wrote a great book on using pandas effectively. It’s called Effective Pandas
Could have gotten a better deal for Crazy Eddie
Well, different humans have different capabilities and we have to choose which one we want to talk to depending on what we are trying to accomplish. Therefore these tools are as good as or better than humans in every possible way. Stop being a hater. ;)
And what’s the point you’re trying to make? If Russia was trying to drop bombs on DC and the US was trying to drop bombs on Moscow, would you think it strange for the US to try to prevent the bombs from being dropped on DC?
I am shocked to learn that Gen AI does not produce reliably accurate information! Shocked, I say!
‘ I always struggle a bit with I'm asked about the "hallucination problem" in LLMs. Because, in some sense, hallucination is all LLMs do. They are dream machines. ...’
Maybe his target audience is anyone who might have read (or written) this??? https://openai.com/index/introducing-openai-o1-preview/
- "A new series of reasoning models for solving hard problems. Available now." - "They can reason through complex tasks and solve harder problems than previous models in science, coding, and math." - "In a qualifying exam for the International Mathematics Olympiad (IMO), GPT-4o correctly solved only 13% of problems, while the reasoning model scored 83%." - "But for complex reasoning tasks this is a significant advancement and represents a new level of AI capability." - "As part of developing these new models, we have come up with a new safety training approach that harnesses their reasoning capabilities to make them adhere to safety and alignment guidelines. By being able to reason about our safety rules in context, it can apply them more effectively. " - "These enhanced reasoning capabilities may be particularly useful if you’re tackling complex problems in science, coding, math, and similar fields."
there are a few more in that post, but clearly OpenAI is pushing the reasoning thing A LOT
Remember. How. It works. Please, please remember how it works. It is generating an answer anew, every single time. It is amazing how often it produces a correct answer, but not at all surprising that it produces inconsistent and sometimes incorrect answers.
1. I believe you that it works well
2. Plausible != Correct - ie, would someone notice if they were reading a slightly different but still logical story?
“Almost perfectly”
It is freaking amazing that this works almost perfectly. Seriously, it’s mind blowing. The problem is, when you keep in mind how the technology works, you realize that the “almost” can never be removed. That’s fine for some use cases but not for others. I understand that human translators make mistakes but they have a conception of truth and correctness, that matters.
Pro tip: on most pumps, the second button from the top on the right side mutes the volume.
This is exactly it! I’m not even that smart and I know that these LLM can never get to the point of being 100% accurate, it’s inherent to how they work. So if I know that, certainly all of the experts working on these systems know that. So many mendacious f*cks.
For Christ’s sake, the guy spells Garry with two ‘r’s, did y’all expect that he would NOT turn out to be a lunatic???
“A cryptographic hash should uniquely identify a file.”
This is not true. It’s not possible to guarantee this. One can be certain that two files DON’T match if they have different hashes, but one cannot be certain that two files DO match based ONLY on the fact that they have the same hash.
I’m quite sure that the OP DOES realize this.
I don't think either one is a contraction.