becquerel

Yeah. To me it seems very intuitive that humor is one of those emergent capabilities that just falls out of models getting more generally intelligent. Anecdotally this has been proven true so far for me. Gemini 2.5 has made me laugh several times at this point, and did so when it was intending to be funny (old models were only funny unintentionally).

2.5 is also one of the few models I've found that will 'play along' with jokes set up in the user prompt. I once asked it what IDE modern necromancers were using since I'd been out of the game for a while, and it played it very straight. Other models felt they had to acknowledge the scenario as fanciful, only engaging with it under an explicit veil of make-believe.

IIRC the guy who makes Aider (Paul Gauthier) has some videos along these lines, of him working on Aider while using Aider (how meta).

If everyone hated AI images, nobody would be creating them.

It crushes the orphans very quickly, and on command, and allows anyone to crush orphans from the comfort of their own home. Most people are low-taste enough that they don't really care about the difference between hand-crushed orphans and artisanal hand-crushed orphans.

You beat me to that idea, haha. I was making an aider for fiction, but your project looks way more useful than what I had.

All labor is bad.

If they do, then I would agree with them.

Aider has built-in functionality for this. You can pass it a command (dotnet test or whatever) and it will autorun it after AI edits. If tests fail it can paste the output into the context for you.

I regret to inform you that war does in fact happen. In fact it happens quite frequently

This is the worst kind of pedantry. There is now code in a text file that wasn't there before. It doesn't really matter to me if they code came from human fingers, an auto generator tool, an LLM or a ouija board. It looks like written code and it compiles like written code - it's written code. You can rightly criticise the use of LLMs in many ways but it is more useful to focus on the actual reasons they cause harm than to set red lines over language.

Roleplaying with robots is currently one of the bigger use-cases for LLMs, esp. across younger generations. Look at the usage stats for something like character.ai. So that is a very clear case where people want to have conversations with computers.

Pretty much where my thoughts on this are. I rarely feel any particular sense of connection to the actual author when I read their books. And I have taken great pleasure from some AI stories (to the degree I put them up on my personal website as a way to keep them around).

I think the doom() function is still getting rolled out across the major browsers. IIRC newer versions of Chrome have a working implementation.

There is truly no suggestions of change to people's lives small enough that someone won't make opposing it part of their identity.

Excel is not merely Turing-complete, but a step above, able to achieve dark things best left unsaid.

I don't know about this specific technique, but I have found it useful to add a line like 'it's OK if you don't know or this isn't possible' at the end of queries. Otherwise LLMs have a tendency to tilt at whatever windmill you give them. Managing tone and expectations with them is a subtle but important art.

Yes, I enjoy it an awful lot. It's my element.

Maybe I am just way deeper in this space that any well-adjusted person should be, but the line of 'did you know LLMs are bad with niche factual information tasks in non-verifiable domains?' has become extremely boring to me. It feels very hard to find something actually new to say on the topic. I find it amazing people still feel the need to talk about it. But then again, I guess most people don't know the difference between a 4o and an R1.

I despise Blow but this is a critique that borders on the incoherent. You could just as well say that when you figure out how to hurt enemies without taking damage, you're done with action games.

And yet LLMs tend to in fact be very funny, just very very rarely on purpose.

"On your fucking telephone. Get real!"

This has been part of LLM usage since day 1, and I say that as an ardent fan of the tech. Let's not forget how much ink has been spilled over that fact that "think through this step by step" measurably improved/improves performance.

Every LLM web app I have used has a disclaimer along these lines prominently featured in the UI. Maybe the disclaimer isn't bright red with gifs of flashing alarms, but the warnings are there for the people who would pay attention to them in the first place.

My natural instinct when I encounter this kind of person is to lavish scorn on them, but I have to hold myself back from that. Partly because it's the mature thing to do, and partly because the coming years will be hard enough on them anyway.

As a sidenote, it amuses me that this author specifically calls out LLMs for being non-deterministic as a grave sin. Despite it being untrue (they've never heard of temperature, I guess), it's funny to me that it should be asserted as a prime directive for computing. Determinism is certainly a useful property where you can get it, but there's a whole lot of useful systems where you can't get it.

That's totally fair. The tech is only barely coming out of the enthusiast adopter phase and there's not a critical mass of content on there to keep most people putting on the headset daily.

That and the ergonomics do still suck, even if I've mostly gotten used to them.

I do think VR will make it, though - starting with the kids. Apparently Gorilla Tag broke 1.5 million players recently, and those are mostly under-15s. The next generation is going to have a strange relationship with computers.

I will personally say that if you ever get the chance, definitely try a Waymo. I did recently for the first time and it's a hell of an experience. You can very vividly imagine it being the future.

I'm also going to stand up for AR/VR here. I'm in a long-distance relationship and me and my partner spend an hour or so in VRChat around two to three times a week. The power that has to reduce the badness of an LDR is well well well well worth the three hundred bucks I paid for a Quest. That and some of the golf games on it are fun.

Fun. Fun! I find it a lot of fun to have a computer spit out pixels based on silly ideas I have. It is very amusing to me

As they say explicitly in the article, this is like criticizing the first rocket to reach the edge of space for not getting anywhere useful.

I am not sharing a platform with people who want me dead.

That feel when a kid has better VCS practices than some people who get paid to develop software...