I love Mistral and I use it whenever I don't care about quality. For example, it's my go to model for generating git commit messages by leveraging their free codestral model.
I don't use skills unless I have something specific to tell the agent. For example, if I want the agent to use Tailwind V3 instead of V4, I'll have a skill for that. Or, if I want the agent to always use the repository pattern for database access, I'll create a skill for that.
I don't need to manage skills files because I have so few of them and they're only a couple lines long.
It's not true at all. I recently vibe coded a REST API in Golang even though I never wrote any production Go code. However, I had to make copious use of my previous experience as an engineer to achieve quality.
You can see this in the real world. The people who are getting the most out of AI are the ones who already have experience coding without AI.
CEOs don't want to admit it but AI can't code anything non-trivial on its own.
I rewrote a very slow backup lambda from Python, which I know well, to Scala, which I have done like three sections of the intro to scala exercises, in about 4 hours, totally vibe coding, but I know enough to skim it and see it is doing things a sensible way, and to get error handling and retries in the robust way. (New job they use Python for slow stuff and Scala for higher performance things).
I agree with you, but now we have managers pushing code to production, people that never coded but are "IT versed"..so you get PR/MR not just with code, but with PR/MR comment written by Claude, Commit Message written by Claude.. all super verbose, and you know the people operating Claude, have no clue. Super dangerous combination. But part of the reality.
Managers are pushing non-trivial code that experienced engineers shouldn't be writing anyway.
If you're worried about not being able to write git commit messages anymore, then you might have been in the wrong career in the first place.
The whole point of AI is to reproduce things that used to be mass produced. AI isn't generating high quality compilers or OS kernels, for example, because we haven't found a way to mass produce those yet.
I suggest you stop listening to clowns like Elon Musk and Super Mario.
> Managers are pushing non-trivial code that experienced engineers shouldn't be writing anyway.
non-trivial code shouldn't be written by experienced engineers? Please explain
> If you're worried about not being able to write git commit messages anymore, then you might have been in the wrong career in the first place.
How did you come to this conclusion? I am concerned about verbose AI documentation, 10 lines of commit message, MR comments with 50+ lines, the typical AI workflow that we all love: One huge commit with 10000+ changes.
> The whole point of AI is to reproduce things that used to be mass produced.
In your opinion, you mean? Because AI is doing everything which involves thinking. Art, Code, Music.. not "mass producted" things..
> I suggest you stop listening to clowns like Elon Musk and Super Mario.
Sounds like you are not really reading, thinking, writing.. only writing i guess?
AI is creating Art, Code, and Music, but only boring stuff that nobody cares about anyway.
Sure, AI can generate your logo, or mix some music that you can play at a party, but at some point, people will call you out for feeding them garbage.
That's when you'll have to hire an experienced graphic designer to design a proper logo, and an experience DJ to create a proper mix based on your audience's mood.
To me AI is just like an IDE. It does stuff that I don't want to do so that I get to work on the fun stuff. Huge win.
I don't find it jaw-dropping because the idea is quite simple. Just feed the model an enormous amount of unethically sourced data, build data centers that cause droughts, and use all chips available so that normal people can't afford to buy RAM anymore.
We're all sacrificing great things in order to make these models more capable. Whether it'll all be worth it, only time will tell.
The arguably sketchy means used to get there —and the resulting side-effects— do not take away from the impressiveness of the emerging capabilities themselves.
I agree about the uncertainty regarding the value for humanity in the long-term. But that's not my point. Being jaw-dropped != being happy and cheering for it.
It's certainly a great thing that AI is improving (I use AI to write code daily myself), but I wouldn't be calling it impressive because it's essentially all about money and government backing.
It’s not human surpassing intelligence though in my opinion. If we had the same constraints on that task then we would also come up with the idea of creating our own viewer and if, like the LLM, the knowledge to do so was embedded in our brain we would do it, it would just take longer because we can’t type as fast. So instead we would buy or compile a viewer since the problem is already solved.
IDK the models definitely do write qualitatively better prose (and better code) than the majority of people. If your definition of "human intelligence" is baseline performance of 0.2% of humanity it's certainly something else.
Better code and better prose are not great indicators of intelligence. Can it help you survive in the middle of the jungle? Can it grow rice? Can it give you first aid? Intelligence is the ability to learn skills and adapt them to the situation. The importance of skills is contextual. Better prose and code is just education, not intelligence.
Imagine it's 2006, there's no LLM in sight and you're arguing on an Internet forum that growing rice is a better highlight of intelligence than computer programming or writing.
Also it absolutely is not education, as repeated failure to educate programmers without aptitude or drill essay-writing into general population has shown.
Then you'll also says that a printer is more intelligent than most of humanity because it can replicate a drawing better than any painter. Or that the ipod is a better musician, etc,...
I can't be amazed at tools. I only applaud the minds that create them. Someone that find a way to grow rice where everyone thought was impossible is way more worthy than whatever opus produced. Just like someone winning the Tour de France is more remarkable than the latest AI benchmark or whatever.
There's likely about a billion cyclists and most of them stand no chance at winning Tour de France ever in their life. And it's funny how we moved from growing rice towards inventing the agriculture.
And look I don't know how to grow rice but I grew other staple foods before. I also wrote substantial texts and programmed computers for decades. Saying that farming is more intellectually challenging than writing or programming is plain ridiculous and counterproductive to whatever argument you were trying to make.
> Saying that farming is more intellectually challenging than writing or programming is plain ridiculous and counterproductive to whatever argument you were trying to make.
I’m not saying that. You invented it on your own.
What I’ve been saying is that writing and programming are not the sole indicators of intelligence. There are plenty of other skills that highlight the intelligence of people.
I don’t say that my compiler is intelligent because it can take my C89 code and optimize it to run on the latest architecture. I also don’t know the latest optimizations techniques. But I would credit the people that are working on GCC and the fact that they know more than me on the subject.
Saying that LLM are better than most humans at programming is like saying that calculators are better at math than most humans, or that a car is faster than humans. It’s a tool, and tools are created. The ones that are creating it are the ones deserving the praise, not the tool.
Speaking of inventing stuff, I'm not sure where you find me praising anything. It was statement of the fact, exactly as "a car is faster than a human".
they don't, cite one example of well written LLM prose? they can definitely infer a lot more slop than people have patience to, and anyone writing or reading slop prose has realised this and replaced sitting around writing and reading slop wiith llm's writing and reading slop and then grug-brains paste 10k words of slop say "grok, pls summarize, no mistakes" and grok tells them "What a brilliantly written article! ... "
Have you ever received an email from trades folk, in the days before LLMs? I'm not saying it's a better writer than Hemingway but very likely better than most your classmates.
We have certainity it will be used to harm most people in the short term while pointificating about long term future. CEO class is very open with their vision and goals, none of them spell anything positive for us.
It seems very unlikely to me that most people will lose either their current job in the short term, or their ability to find work at all in the medium term.
It does seem like something to worry about for a small percentage of people in the short term (just as the China shock hurt a number of manufacturing workers in the 90s/00s) and maybe a majority of people when AGI exists, but who knows when that will be.
I’m simply not seeing the certainty you’re seeing. There are lots of risks associated with power concentration, but I’m not seeing evidence of that yet. What I am seeing are open weights models, and even just competition between closed weight model providers alone, turn these systems into commodity products. Additionally I am not seeing any kind of massive labor demand destruction, even though we’re nearly 6 months into the middle of a global energy supply shock. Maybe your fears are realized eventually but I do not think (a) they are certain or (b) they will happen in the near term.
Exactly. I'm a nerd too and we nerds can real hypocrites because we spend too much time alone. I will happily turn a blind eye to world suffering if it means I get to work less but get paid the same.
AI CEOs aren't different. At some point we'll have to admit that we're sacrificing others for our own benefit.
How about total surveillance total thought control total power in a hands of few trillionaires in the near future. Drones killing undesirables with cheering crowds watching.
It is not like it is much different right now but it may be significantly worse.
And even that is one of the “good” outcomes assuming no ASI.
That is a bad outcome. But there are better possible outcomes that the doomers ignore. How about AI curing cancer? I think that alone is worth the risks.
By the time we spend all the earths resources to train AI models so that cancer is cured, we could have spent those same resources training real humans. The literacy rate of undeveloped countries is still abysmal. We could train more people and fund more research.
And humans can keep growing while AI can only grow if it has access to high quality training data. Who provides the data? Humans.
Then again, looking back, that’s the way history is made. Modern democracies in the west are built on foundations of slavery and colonialism; yet we cherish them as to all the greatness they have brought as well.
We need some amount of tolerance to ambiguity, and LLMs are undoubtedly both extremely cool but also extremely concerning.
Edit: Note that I am not saying that this is good, or desirable; just that it is. The technology is here know and not going anywhere, with all the flaws of its conception. We can be skeptical and curious at the same time.
The west thrived by leveraging slavery and colonialism at the expense of everyone else. Then democracies were built by moving away from both slavery and colonialism so that the world can grow as a whole.
It's possible that we'll have to move away from AI at some point in order for society to continue growing.
The ottoman empire leveraged slavery and colonialism just as much, and there's no progress to be seen from them, at least not when compared to the west.
It's the renaissance and the scientific and industrial revolutions, that's what gave the west the opportunity to thrive.
Unlike western powers, The Ottoman empire didn't use slavery and colonialism to create capital. Instead, it wanted slaves so that it could grow its military.
The west also made its goal to create a new world order while the Ottoman empire's goal was mostly revenue extraction. The British systematically deindustrialized India’s textile sector to turn it from a competitor into an exporter of raw cotton and an importer of Lancashire cloth. The Ottoman would opt for taxation instead of usurpation of an entire industry.
Scientific revolutions was already happening during the bronze age. The west simply leveraged existing systems while making use of violence and exploitation to rise to the top.
Please tell me you're not so delusional to think that slavery or colonialism is a uniquely "western" trait. Empires, nation-states, and groups all sought to expand; some, for various reasons, did much better than others.
I wouldn't pay for it though.