In case you haven’t seen it, the paper is here - https://machinelearning.apple.com/research/illusion-of-thinking (PDF linked on the left).

The puzzles the researchers have chosen are spatial and logical reasoning puzzles - so certainly not the natural domain of LLMs. The paper doesn’t unfortunately give a clear definition of reasoning, I think I might surmise it as “analysing a scenario and extracting rules that allow you to achieve a desired outcome”.

They also don’t provide the prompts they use - not even for the cases where they say they provide the algorithm in the prompt, which makes that aspect less convincing to me.

What I did find noteworthy was how the models were able to provide around 100 steps correctly for larger Tower of Hanoi problems, but only 4 or 5 correct steps for larger River Crossing problems. I think the River Crossing problem is like the one where you have a boatman who wants to get a fox, a chicken and a bag of rice across a river, but can only take two in his boat at one time? In any case, the researchers suggest that this could be because there will be plenty of examples of Towers of Hanoi with larger numbers of disks, while not so many examples of the River Crossing with a lot more than the typical number of items being ferried across. This being more evidence that the LLMs (and LRMs) are merely recalling examples they’ve seen, rather than genuinely working them out.

I think it’s an easy mistake to confuse sentience and intelligence. It happens in Hollywood all the time - “Skynet began learning at a geometric rate, on July 23 2004 it became self-aware” yadda yadda

But that’s not how sentience works. We don’t have to be as intelligent as Skynet supposedly was in order to be sentient. We don’t start our lives as unthinking robots, and then one day - once we’ve finally got a handle on calculus or a deep enough understanding of the causes of the fall of the Roman empire - we suddenly blink into consciousness. On the contrary, even the stupidest humans are accepted as being sentient. Even a young child, not yet able to walk or do anything more than vomit on their parents’ new sofa, is considered as a conscious individual.

So there is no reason to think that AI - whenever it should be achieved, if ever - will be conscious any more than the dumb computers that precede it.

While true, why are you linking this comment in almost all the other comments?

I’ve been stuck repeatedly asking myself this question ever since reading your comment 😩 Please be careful about throwing infinite while true loops around! Now I need someone to Ctrl-C me.

I’m probably misunderstanding as I rarely use word processing software, so I apologise if you talking about something more than the system’s own handling of touchpad scrolling! here’s the settings applet for XFCE, I think every DE will have similar options (it does even offer circular scrolling, but I know you aren’t looking for that):

Gesture scrolling? You mean like making clockwise or anticlockwise circles to scroll up or down? I’d have thought that kind of functionality would be handled by the touchpad driver, not individual programs.

a sign of utter desperation on the human’s part.

Yes it seems to be the same underlying issue that leads some people to throw money at only fans streamers and such like. A complete starvation of personal contact that leads people to willingly live in a fantasy world.

It was also a true AI wasn’t it? It ran locally and was never turned off, so conversations with it were private and it continued to “exist” and develop by itself.

Here’s a list of well-supported USB WiFi adapters - https://github.com/morrownr/USB-WiFi/blob/main/home/USB_WiFi_Adapters_that_are_supported_with_Linux_in-kernel_drivers.md

Can you just stream video and audio directly, like a standard IP camera? This list of solutions in the Raspberry Pi documentation could have some ideas - https://www.raspberrypi.com/documentation/computers/camera_software.html#stream-video-over-a-network-with-rpicam-apps (there are some RPi specific solutions, but also general Linux approaches e.g. ffplay)

I think they mean that ARM became dominant by widely licencing its RISC architecture to pretty much anyone. This startup wants to make RISC V designs and licence them to various chip manufacturers - so they won’t be in the business of making chips themselves, just the design.

But as long as they are RISC V chips, then they would run the same software as any other RISC V chips.

Would that be a risk? Isn’t the whole point of RISC V that its ISA is open and free to use? That’s not the case for ARM or Intel’s x86 architecture.

In case you don’t already know about it, paccache (part of the pacman-contrib package) will let you easily remove old packages from the pacman cache

fraid I generated a tl;dr for this rather verbose article:

“Home directories are a mess because too many apps ignore XDG spec and dump dotfiles everywhere. The problem isn’t just legacy software—new apps do it too, often out of ignorance or laziness. Windows has similar issues with profile folders. Fixing it requires devs to actually follow standards, but many resist due to inertia or ‘my way is better’ thinking. Users should push back and demand proper XDG compliance to keep $HOME clean.”

Well like it or not, your footer is just a part of your comments, and so people are invited to respond however they wish when you post it on lemmy. If you don’t like people making the same replies, you can simply stop posting the same content in every comment.

artistic licence innit - based being the opposite of cringe.