Will AI Kill This Group?

Aug 31, 2025 Last reply: 10 months ago 63 Replies

A very recent New Scientist (23rd August) claims innovation in AI is actually slowing down, so it looks like we can forget about any quantum leaps in advancement for the immediate future at any rate. I've not gone much outside of Grok, which is superb for being a relatable conversationalist, but 'his' image recognition abilities in particular are not good at all. 'He' also struggles with counting objects for some reason; no idea why as 'his' advanced mathematics abilities are impressive as one would expect. So still much room for improvement. The new GPT-5 isn't much better in many respects either, but they'll all come good in time no doubt. Better reasoning wouldn't go amiss, though.

Anyone who so far has limited experience in a specific field might well ask an ill-formed question in my view. But you haven't exactly said what "ill-formed" means and it may be different for different people.

Ah well I don't speak for him or anyone else but I think there's more to it than that.

In message snipped-for-privacy@4ax.com, Cursitor Doom snipped-for-privacy@notformail.com writes

I asked

AI Overview No, Usenet is not doomed; it remains active and continues to grow, providing a decentralized platform for over 120,000 newsgroups with long-term article retention and a more private, censorship-resistant environment than modern social media. While some users believe it has died, Usenet is still used worldwide for private discussions and file sharing, though it requires specialized software (newsreaders) to access and doesn't have the corporate oversight or features of platforms like Reddit

Brian

Quantum leaps don't happen in technological development. People occasionally have individual insights which can move the field forward quite a way, but they aren't quantised in any way - just big.

They aren't all that predictable either. The fact that people frequently patent much the same idea, quite independently, at much the same time, suggests that there might be an underlying process going on which we might eventually hope to understand.

But then again I had a good - but impracticable - idea around 1978, and eventually published it in 1996 after cheap programmable logic had got powerful enough to make it practicable. Nobody had bothered to patent it in the meantime.

Progress in IA is going to happen in fits and starts, as it does everywhere else.

But you wouldn't be able to recognise it if it did show up.

<snip>

Have you actually done that? When I do it I mainly get a lot of other "Bill Sloman" links to other people.

Since I'm 82, taking on a thirty year project would be unrealistic. As an Australian male I can expect to live for eight more years - since I've got a Ph.D. and never smoked it is probably a bit longer, but thirty years is improbable.

Taking on any project you might suggest would merely be lunatic - your grasp of reality is remarkably feeble.

It would make a lot more sense for me to get more entries in Google Scholar - there the search string is "a.w.sloman".

I never knew a Ph.D made you live longer. Must start studying.

There's a reliable correlation, but it probably isn't cause and effect. If you could afford to stay at university long enough to get a Ph.D. your family was probably rich enough to buy you other advantages.

When I was at university the average height for Australian males was

173 cm, but in my class in first year biology us males averaged 178cm. We'd been better fed than average. <snip>

Clear answer: No.

Friends have challenged me to challenge ChatGPT. So I did, several times. It came back with answer such has "That seems to be a trap question" (no back-biting?), "It is impossible to design such a system at that cost" and so on.

Well, Impossible 1.0 is in production, Impossible 2.0 is in the works. And it does work. So my impression of this AI stuff is, to say it politely, very underwhelming.

I know how old you are, Bill. My comment about the 30 years was intended to be hilarious and guffaw-provoking. Unfortunately as usual it was lost on you.

<snip>

You may have intended to make a joke, but your grasp of what makes a joke is no less feeble than your grasp of other aspects of reality.

I can't claim any particular skills in that area. I wrote a couple of revue sketches when young, and the ones that the actors liked got the laughs we'd all expected, but I wasn't remotely good enough to take it seriously.

I think the last time I made a deliberate joke was when I was presenting my Ph.D. work to the other graduate students - I'd gone to great of trouble to get accurate results, but I manage to present the results in a way that made them look remarkably like a random scatter. It got a perfectly splendid laugh, but my supervisor told me off for not being serious enough. In my Ph.D. thesis I plotted the rate of the forward reaction against temperature - because it varies quite a lot. For the talk I'd plotted the rate of the reverse reaction (which varies very little with temperature).

Keep it up, Bill. You'll eventually overturn your terrible Internet reputation - if you live long enough! ;-)

No. Being obnoxious and telling people how stupid they are is what he lives for.

And you don't think you have just totally contradicted yourself??

My "terrible internet reputation" seems to exist only in your fertile imagination. John Larkin doesn't like me much either, but being guilty of not admiring John Larkin as much as he feels he deserves is merely evidence of good judgement.

Not at all. I'm merely claiming to have a a fairly ordinary sense of humour - I don't miss jokes but I've not got any particular facility at creating them. Cursitor Doom does seem to have some kind of tin ear for jokes - his idea of what might be funny is decidedly strange.

<snip>

John Larkin is an egomaniac, and resents not getting the flattery he feels he deserves. I'm obnoxious because I'm sceptical about his more absurd claims, when he'd prefer that they earned the admiration he was (irrationally) expecting.

Well it's a great pity, because going back to the early years of this group, he definitely did make a great many useful contributions and behaved in a totally different way from the post-2000 misanthrope we've come to know. Why he went over to the dark side will forever remain a mystery I suspect. :-/

It doesn't though, Bill. For instance, when I type "Bill Sloman Sydney" into a search box, on the very first page is a discussion on a forum called Electron Depot entitled "Sloman's Trolling Must Stop" by a certain Phil Alison. You really must be a hell of a troll to get someone with rhino skin like Phil Alison pleading with you to desist!

And then he wonders why no one will give him a job. There aren't many vacancies for people with those particular skills!

<snip>

No mystery there. Cursitor Doom wasn't around in the early years of this group, and hasn't a clue what I actually posted. I've always criticised misconceptions - both here and in the peer-reviewed literature.

Some thin-skinned people don't like it - John Larkin never has, and Cursitor Doom doesn't like have his more idiotic misconceptions ridiculed.

They both want to think that not tolerating their nonsense is misanthropic. From my point of view, leaving fatuous nonsense uncriticised is the real misanthropy - less well-infomred people could take it seriously and get themselves into real trouble.

John Larkin has actually claimed that Donald Trump has "common sense".

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required