ChatGPT Is a Gimmick

One of the realisations I've had recently is that the AI hype feels like another level from what's come before because AI itself is creating the "hype" content fed to me (and my bosses and colleagues) all over social media.

The FOMO tech people are having with AI is out of control - everyone assumes that everyone else is having way more success with it than they are.

I use LLMs to check solutions for graduate level math and physics problem I'm working on. Can I 100% trust their final output? Of course not, but I know enough about the domain to tell whether they discovered mistakes in my solutions or not. And they do a pretty good job and have found mistakes in my reasoning many times.

I also use them for various coding tasks and they, together with agent frameworks, regularly do refactoring or small feature implementations in 1-2 minutes that would've taken me 10-20 minutes. They've probably increased my developer productivity by 2-3x overall, and by a lot more when I'm working with technology stacks that I'm not so familiar with or haven't worked with for a while. And I've been an engineer for almost 30 years.

So yea, I think you're just using them wrong.

There are absolutely times when one can get LLMs to "say something valuable". I am still learning how to put them to good use, but here are some areas where I have found clear wins:

* Super-powered thesaurus

A traditional thesaurus can only take a word and provide alternative words; with an LLM, you can take a whole phrase or sentence and say: "give me more ways to express the same idea".

I have done this occasionally when writing, and the results were great. No, I do not blindly cut-and-paste LLM output, and would never do so. But when I am struggling to phrase something just right, often the LLM will come up with a sentence which is close, and which I can tweak to get it exactly the way I want.

* Explaining a step in a mathematical proof.

When reading mathematical research papers or textbooks, I often find myself stuck at some point in a proof, not able to see how one step follows from the previous ones. Asking an LLM to explain can be a great way to get unstuck.

When doing so, you absolutely cannot take whatever the LLM says as 'gospel'. They can and will get confused and say illogical things. But if you call the LLM out on its nonsense, it can often correct itself and come up with a better explanation. Even if it doesn't get all the way to the right answer, as long as it gets close enough to give me the flash of inspiration I needed, that's enough for me.

* Super-powered programming language reference manual

I have written computer software in more than 20 programming languages, and can't remember all the standard library functions in each language, what the order of parameters are, and so on.

There are definitely times when going to a manpage or reference manual is better. But there are also times when asking an LLM is better.

These “AI is a gimmick that does nothing” articles mostly just communicate to me that most people lack imagination. I have gotten so much value out of AI (specifically ChatGPT and Midjourney) that it’s hard to imagine that a few years ago this was not even remotely possible.

The difference, it seems, is that I’ve been looking at these tools and thinking how I can use them in creative ways to accomplish a goal - and not just treating it like a magic button that solves all problems without fine-tuning.

To give you a few examples:

- There is something called the Picture Superiority Effect, which states that humans remember images better than merely words. I have been interested in applying this to language learning – imagine a unique image for each word you’re learning in German, for example. A few years ago I was about to hire an illustrator to make these images for me, but now with Midjourney or other image creators, I can functionally make unlimited unique images for $30 a month. This is a massive new development that wasn’t possible before.

- I have been working on a list of AI tools that would be useful for “thinking” or analyzing a piece of writing. Things like: analyze the assumptions in this piece; find related concepts with genealogical links; check if this idea is original or not; rephrase this argument as a series of Socratic dialogues. And so on. This kind of thing has been immensely helpful in evaluating my own personal essays and ideas, and prior to AI tools it, again, was not really possible unless I hired someone to critique my work.

The key for both of these example use cases is that I have absolutely no expectation of perfection. I don’t expect the AI images or text to be free of errors. The point is to use them as messy, creative tools that open up possibilities and unconsidered angles, not to do all the work for you.

I think it's starting to change.

I'm an AI sceptic (and generally disregard most AI announcements). I don't think it's going to replace SWE at all.

I've been chunking the same questions both to Gemini and GPT and I'd say about until ~8 months ago they were both as bad as each other and basically useless.

However, recently Gemini has gotten noticeable better and has never hallucinated.

I don't let it write any code for me. Instead I treat Gemini as a 10+ YoE on {{subject}}.

Working as platform engineer, my subjects are broad so it's very useful to have a rubber duck ready to go on almost any topic.

I don't use copilot or any other AI. So I can't compare it to those.