All of these anecdotal stories about "LLM" failures need to go into more detail about what model, prompt, and scaffolding was used. It makes a huge difference. Were they using Deep Research, which searches for relevant articles and brings facts from them into the report? Or did they type a few sentences into ChatGPT Free and blindly take it on faith? LLMs are _tools_, not oracles. They require thought and skill to us…
Why do you think these details are important? The entire point of these tools is that I am supposed to be able to trust what they say. The hard work is precisely to be able to spot which things are true and false. If I could do that I wouldn't need an assistant.
Because then I can know whether the hallucinations they encountered are a little surprising, or not surprising at all.
I get so confused on this. I play around, test, and mess with LLMs all the time and they are miraculous. Just amazing, doing things we dreamed about for decades. I mean, I can ask for obscure things with subtle nuance where I misspell words and mess up my question and it figures it out. It talks to me like a person. It generates really cool images. It helps me write code. And just tons of other stuff that astounds me…
Look man, and I'm saying this not to you but to everyone who is in this boat; you've got to understand that after a while, the novelty wears off. We get it. It's miraculous that some gigabytes of matrices can possibly interpret and generate text, images, and sound. It's fascinating, it really is. Sometimes, it's borderline terrifying. But, if you spend too much time fawning over how impressive these things are, you m…
>they can dump out a boilerplate react frontend to a CRUD API
This is so clearly biased that it boarders on parody. You can only get out what you put in.
The real use case of current LLMs is that any project that would previously require collaboration can now be down solo with a much faster turnover. Of course in 20 years when compute finally catches up they will just be super intelligent AGI
I get so confused on this. I play around, test, and mess with LLMs all the time and they are miraculous. Just amazing, doing things we dreamed about for decades. I mean, I can ask for obscure things with subtle nuance where I misspell words and mess up my question and it figures it out. It talks to me like a person. It generates really cool images. It helps me write code. And just tons of other stuff that astounds me…
Indeed, it is the stuff of science fiction, and the you get an "akshually, it's just statistics" comment. I feel people projecting their fears, because deep down, they're simply afraid.
One of the things that’s hard about these discussions is that behind them is an obscene amount of money and hype. She’s not responding to realists like you. She’s responding to the bulls. The people saying these tools will be able to run the world by the end of this year, maybe next. And that’s honestly unfair to you since you do awesome realistic and level headed work with LLM. But I think it’s important when having…
Possibly a reaction to Bill Gates recent statements that it will begin replacing doctors and teachers. It's ridiculous to say LLMs are incredibly useful and valuable. It's highly dubious to think they can be trusted with actual critical tasks without careful supervision.
It's honestly so scary because Sam Altman and his ilk would gladly replace all teachers with LLM's right now, because it makes their lines go up, doesn't matter to them that would result in a generation of dumb people in like 10 years. Honestly would just create more LLM users for them to sell to so its a win win I guess, but it completely fucks up our world.
And that says… what? The entire LLM technology is worthless for all applications, from all implementations? A company I worked for spent millions on a customer service solution that never worked. I wouldn’t say that contracted software is useless.
I agree. I use LLMs heavily for gruntwork development tasks (porting shell scripts to Ansible is an example of something I just applied them to). For these purposes, it works well. LLMs excel in situations where you need repetitive, simple adjustments on a large scale. IE: swap every postgres insert query, with the corresponding mysql insert query. A lot of the "LLMs are worthless" talk I see tends to follow this pat…
>swap every postgres insert query, with the corresponding mysql insert query.
If the data and relationships in those insert queries matter, at some unknown future date you may find yourself cursing your choice to use an LLM for this task. On the other hand you might not ever find out and just experience a faint sense of unease as to why your customers have quietly dropped your product.
I become more and more convinced with each of these tweets/blogs/threads that using LLMs well is a skill set akin to using Search well. It’s been a common mantra - at least in my bubble of technologists - that a good majority of the software engineering skill set is knowing how to search well. Knowing when search is the right tool, how to format a query, how to peruse the results and find the useful ones, what result…
LLMs function as a new kind of search engine, one that is amazingly useful because it can surface things that traditional search could never dream of. Don't know the name of a concept, just describe it vaguely and the LLM will pull out the term. Are you not sure what kind of information even goes into a cover letter or what's customary to talk about? Ask an LLM to write you one, it will be bland and generic sure but that's not the point because you now know the "shape" of what they're supposed to look like and that's great for getting unblocked. Have you stumbled across a passage of text that's almost English but you're not really sure what to look up to decipher it? Paste it into the LLM and it will tell you that it's "Early Modern English" which you can look up to confirm and get a dictionary for.
My experience (almost exclusively Claude), has just been so different that I don't know what to say. Some of the examples are the kinds of things I explicitly wouldn't expect LLMs to be particularly good at so I wouldn't use them for, and others, she says that it just doesn't work for her, and that experience is just so different than mine that I don't know how to respond. I think that there are two kinds of people w…
I think you have an interesting point of view and I enjoy reading your comments, but it sounds a little absurd and circular to discount people's negativity about LLMs simply because it's their fault for using an LLM for something it's not good at. I don't believe in the strawman characterization of people giving LLMs incredibly complex problems and being unreasonably judgemental about the unsatisfactory results. I work with LLMs every day. Companies pay me good money to implement reliable solutions that use these models and it's a struggle. Currently I'm working with Claude 3.5 to analyze customer support chats. Just as many times as it makes impressive, nuanced judgments it fails to correctly make simple trivial judgements. Just as many times as it follows my prompt to a tee, it also forgets or ignores important parts of my prompt. So the problem for me is it's incredibly difficult to know when it'll succeed and when it'll fail for a given input. Am I unreasonable for having these frustrations? Am I unreasonable for doubting the efficacy of LLMs to address problems that many believe are already solved? Can you understand my frustration to see people characterize me as such because ChatGPT made a really cool image for them once?
The technology is not just less than superintelligence, for many applications it is less than prior forms of intelligence like traditional search and Stack Exchange, which were easily accessible 3 years ago and are in the process of being displaced by LLMs. I find that outcome unimpressive. And this Tweeter's complaints do not sound like a demand for superintelligence. They sound like a demand for something far more…
"They continue to fabricate links, references, and quotes, like they did from day one." - "I ask them to give me a source for an alleged quote, I click on the link, it returns a 404 error." Why have these companies not manually engineered out a problem like this by now? Just do a check to make sure links are real. That's pretty unimpressive to me. There are no fabricated links, references, or quotes, in OpenAI's GPT…
While I agree, it doesn't stop business folks pushing for its use in area where it is inappropriate. That is, at least for me, part of the skepticism.
All of these anecdotal stories about "LLM" failures need to go into more detail about what model, prompt, and scaffolding was used. It makes a huge difference. Were they using Deep Research, which searches for relevant articles and brings facts from them into the report? Or did they type a few sentences into ChatGPT Free and blindly take it on faith? LLMs are _tools_, not oracles. They require thought and skill to us…
If any non-trivial ask of an LLM also requires the prompts/scaffolding to be listed, and independently verified, along with its output, their utility is severely diminished. They should be saving time not giving us extra homework. Far better to just get these problems resolved.
That isn't what I'm saying. I'm saying you can't make a blanket statement that LLMs in general aren't fit for some particular task. There are certainly tasks where no LLM is competent, but for others, some LLMs might be suitable while others are not. At least some level of detail beyond "they used an LLM" is required to know whether a) there was user error involved, or b) an inappropriate tool was chosen.
I get so confused on this. I play around, test, and mess with LLMs all the time and they are miraculous. Just amazing, doing things we dreamed about for decades. I mean, I can ask for obscure things with subtle nuance where I misspell words and mess up my question and it figures it out. It talks to me like a person. It generates really cool images. It helps me write code. And just tons of other stuff that astounds me…
The problem Sabine tries to communicate is that reality is different from what the cash-heads behind main commercial models are trying to portray. They push the narrative that they’ve created something akin to human cognition, when in reality, they’ve just optimised prediction algorithms on an unprecedented scale. They are trying to say that they created Intelligence, which is the ability to acquire and apply knowled…
LLMs seem akin to parts of human cognition, maybe the initial fast thinking bit when ideas pop up in a second of two. But any human writing a review with links to sources would look them up and check the are they right ones that match the initial idea. Current LLMs don't seem to do that, at least the ones Sabine complains about.
Akin to human cognition but still a few bricks short of a load, as it were.