I've already started my AI UltraSuperDuper Intelligence unit, so they are way behind.

25% still seems a bit high to me. I do wonder if they really have forgotten how to do long division, or simply forgot what the words 'long division' mean. Like if you told them to work a division problem by hand, would they naturally just do long division, forgetting that was all that long division meant?

Because either:

a) It works as intended and the job inherently fast-tracks self-obsolescence.

b) It doesn't work as intended and this job evaporates as the hype money comes back down to earth.

No matter how well/poorly this current technology goes, this is a job that is not set to be a career.

Just like people claiming to be "prompt engineers", either the LLMs work and you are a useless middle man or they don't work and people don't want to fool with you. Just like "webmaster" was a thing just by being able to edit HTML files and that evaporated in the early 2000s.

Even for those that don't need that much range, there can be benefits.

The reason they can tout a goal of 600 mile range is that solid state batteries have much more energy per kg. NMC batteries are roughly 200Wh/kg, *maybe* someone can get 350Wh/kg in the most aggressive marketing claims I could find. Solid state batteries are more like 700-800 Wh/kg.

So if you say for a given car and lifestyle you could accept a 150 mile range, then you could produce for example an electric Miata that could weigh about the same as the ICE miata (ICE miata drivetrain+fuel weighs about 400 lbs, a credible electric motor might weigh 200lbs and a 150 mile solid state might also weigh about 200lbs.) A miata is the sort of car that may likely get away with low range as a 'fun' car you probably don't want to be road tripping in anyway. Or targeting a 300 mile range and being only 200lbs heavier instead of having to be 600 lbs heavier with NMC.

For these people the options are either making Agentic AI able to do everything or them doing nothing at all because they don't actually know how. One of those options includes maybe money for some period of time, and the other has no opportunity for money.

I didn't read much about the other one, but the SaaStr guy was obviously a true believer. He had been making posts gleefully detailing his vibe coding journey and then clearly feeling betrayed by how quickly it all went south out of his control.

The thing is the "vibe coding" movement is about not needing any of the technical skills that would have you actually understand testing/staging, let alone actually making an environment that would actually enforce it to an otherwise enabled "agentic" LLM.

Having another LLM to fix the other LLM is just the blind leading the blind.

It is a solvable issue, but the solutions run counter to the expectations around the immense amount of money in play. LLMs are useful, but not as useful as the unprecedented investment would demand. After the bubble deflates a bit, maybe we will see good utilization of LLMs, but right now there's a lot of high risk grifting in play and a lot of people getting in way more over their head than they formerly could manage.

This is a failure of AI marketing, and how the AI companies encourage this behavior.

There are a *lot* of people without the skillset but have seen the dollars. Either they watch from the outside or they manage to become tech execs by bullshitting other non-tech executives.

Then AI companies talk up just a prose prompt and bam, you have a full stack application. The experienced can evaluate it reasonably in the context of code completion/prompt developing specific function, with a managable review surface and their own experience to evaluate and get a sense for how likely an attempted LLM usage is going to be productive and how much fixing it's going to need. The inexperienced cannot do that, so they make a go at vibe-coding up what would be tutorial fodder. Then they see a hopelessly intimidating full stack application that does exactly what they say and erroneously conclude that it must be generally capable.

So some folks can be happier vibe coding up a shovelware game with pretty low stakes and decent chance of success (though it sucks to dilute the game landscape with too much content that is utterly devoid of creativity). Some people think they can get rich quick by participating in a skilled industry without any skills (The Saastr story is particularly funny, they purport to be a resource for other developers, but can't even develop themselves). Not great, but less of a risk. The real risk are those tech execs high on BS and low on technical acumen, who are generally insecure about people that have an advantage over him. He sees a great equalizer and all his personal sources that could grade it are people he doesn't trust. So it's good to see stories like this for those executives to maybe, possibly understand the risk when they talk about laying off all or nearly all their software developers (yes, a few weeks ago an executive with hundreds of developers told me this was basically his plan, and I was only safe because I understood my respective customer base better than marketing, sales, and the executives, but most of his developers just do what he says and his "executive insight" is valuable, but their work is prime to be replaced by executives just vibe coding up stuff directly instead of having developers do it).

I chose RC boat because some people are using this competing at "stupid human trick" as an example of intrinsic proof of LLMs being able to supersede humans.

In the scenario of an olympic swimming competition, an autonomous boat vs a manned boat would show no difference to each other, both would compete the task much better than a human. It's a useless test to measure general utility. Just like a person swimming a 1500 meter distance is not really a useful indicator on its own of how useful they are. These Math Olympiads are similar in that they are not particularly indicative of people being useful. The same ways that we would stress a human in impressive ways does not mean that a computer coming at it from a different approach should be considered to have broadly superseded the humans.

Yes, a real boat is valuable and LLM can be valuable when utilized correctly, but in the face of exaggerated hype some pessimistic reality check is called for to balance expectations.

LLMs have a great deal of utility and extend the reach of computing to a fair amount of scope that was formerly out of reach to computing, but they don't "think" and the branding of the "reasoning" models is marketing, not substantive.

The best evidence is reviewing so-called "reasoning chains" and how the mistakes behave.

Mistakes are certainly plausible in "true thinking", but the way they interact with the rest of the "chain" is frequently telling. It flubs a "step" in the reasoning and if it were actual reasoning, that should propagate to the rest of the chain. However when a mistake is made in the chain, it's often isolated and the "next step" is written as if the previous step said a correct thing, without ever needing to "correct" itself or otherwise recognize the error. What has been found is that if you have it generate more content and dispose of designated "intermediate" content you have a better result, and the intermediate throwaway content certainly looks like what a thought process may look like, but ultimately it's just more prose and mistakes in the content continue to have an interesting behavior of isolation rather than contaminating the rest of an otherwise ok result.

I think most people can do long division, but almost no one would bother day to day, because why would they?

When the model is used for inference, yes. But I assume he was speaking to the awkwardness of training. Take a machine vision that has never been trained on dogs and cats, feed it a dozen labeled images of cats and dogs to retrain it to add dog/cat recognition. Then try to do inference on that model and it will be utterly useless still for dog/cat recognition. Take a model trained on normal images. Then have it try recognition on a fisheye lense. It will fail because it has no idea. You might hope to retrain it to recognize the fisheye distortion generically, but generally that won't work and you have to retrain it to catch the fisheye variant of everything you want. After retraining, if you just slap the fisheye variant into a normal picture, the model is likely to be oblivious to the anomaly.

Don't know if this is an argument about 'thinking', but it is certainly a difference in 'learning', that AI models need to consume way way more than a human to start to give useful results. A model training to operate a car is still oddly dodgy even after consuming more car operating hours than a human will consume in a lifetime. In areas where we have enough training data to feed, this is fine, but it's certainly different than a human learning, which can do something better to extend less training to better results.

I'm not in denial, the LLMs and other forms of AI have utility, but expectations have to be mitigated.

Was in a discussion with a software executive a couple weeks back who said he fully anticipates he can lay off every one of his software developers and testers in the next year and only have to retain the 'important' people: the executives and sales people.

People see articles like this show how LLMs enable computing to reach another tier of 'stupid human tricks', which is certainly novel, but people overextend exactly what this means. We put too much focus on whether the LLMs can navigate tests explicitly designed to be graded and passable for humans, with effort and imagine that means they are able to chart the less well trodden set of problems that have an unknown or maybe no solution at all.

There's a lot of stuff LLMs can do that computing wasn't able to do before, and that does potentially speak to a great deal of the things humans do today, but we have to temper the wild expectations people are getting.

The point is we have a myriad of "tests that are hard for humans, but don't necessarily translate to anything vaguely useful". In academics, a lot of tests are only demanding of reasoning ability because the human has limited memory. Computers short on actual "reasoning" largely make up for it by having just mind boggling amounts of something more akin to recall than reasoning (it's something a bit weirder, but as far as analogies go, recall is closer).

It's kind of like bragging that your RC boat could get a gold medal in the Olympic 1500 meter freestyle. It didn't complete the challenge in the same way, and that boat would be unable to, for example, save someone that is about to drown, because the boat can just go places, it can't do what that human swimmer could do. That person swimming 1500m itself is a useless feat of interest, and not really directly useful in and of itself.

If it needs USB-PD, looks like the products are pretty pricey.

If it can take the old fashioned USB power, looks like a splitter is reasonably available for like $15.

The commenter clearly seemed to think the world was going to be supremely unfair to the CEO (turns out 'exec' is ambiguous, as the man, his wife, the mistress, and the mistress' husband are all executives one place or another). You said the exec deserves to lose because of his actions, which seems to be inconsistent. The commenter's stance is based on his blatant assumption that the wife was not earning money and the mistress was just some gold digger, and that even if the wife wasn't earning money, that if the split happens it's unfair for her to get a cut of the CEOs wealth that he earned.

The assertions of misogyny are because he filled in the gaps he didn't know with assumptions consistent with negative stereotypes of women in these situations. He jumped right to the fiction of the struggling man paying huge alimony to some indolent ex-wife living a life of luxury. That the mistress was only in it for gold
digging.