Weirdly, most of the work in the world is still performed by people

-- AC, 2024, responding to the idea that automation will take over

Weirdly, most of the transport in the world is still performed by horses

-- Buggy whip manufacturers, 1886, responding to the idea that the IC vehicle will take over

Very simple: Currently known Physics mandates that everything large-scale goes down to just 4 fundamental forces and everything small-scale goes down to quantum-effects. Neither of the two can form consciousness as currently understood.

So far, without exception, that's been an indication of a lack of understanding. Not an indication of impossibility. There are lots of things we don't understand, or fully understand. And many more that we didn't understand, but now understand. That doesn't imply current knowledge gaps must be filled with instances of unknown physics. Particularly as every time we have gained an understanding of something, is has 100% turned out to be depending upon no more than the same physics everything else leverages, or a deeper understanding of same. That's clearly the way to bet; it's not an equal balance between physics and magic.

I argue the question [of non-physics brainops] is open.

Well, it's open in the same way that the question of practical commercial fusion is open. It's an unsolved problem, but the physics appear to be the same physics everything else has turned out to depend upon. Everything we know of, literally everything, works "because physics." You did say "known physics" — which is an interesting hedge — however we've observed no processes in the brain that perform in ways that imply unknown physics, nor have we found any material structures in the brain that cannot be explained by known physics, so there's that.

There is a pretty large gap between "we haven't figured out how this very complex system works" versus "must be magic", and the assumptions on the opposite sides of that gap are not worthy of equal weight.

That is why...

At this time, Physicalism is just as much belief as the opposite.

...is an unjustifiable position at this time. The evidence — all of it, and there's plenty — is that the brain is a system operating within the bounds of physics. Not only that, but within the bounds of known physics.

It's my belief that the basic technologies of the LLMs are sufficient [for AGI]

Well, if you're going with "AGI is a really good LLM", okay, perhaps.

But if you're using AGI to describe a conscious intelligence (you know, what "AI" meant before the marketers starting calling thermostats intelligent), i.e. a synthetic person, no, almost certainly not. LLMs produce streams of words according to probabilities developed from their training data and guided by the input query or queries. This is why some of the word streams, while grammatically coherent, are factually nonsensical: this is misprediction (somewhat risibly called "hallucination" by the marketing types. Protip: it's not hallucination.) Putting words together according to probability is not reasoning. LLMs don't think.

Having said that, an LLM might end up being part of an AGI (and yes, I do mean a synthetic person) but the role is almost certain to be limited to assembling output from a system that actually puts information together in such a way as to actually understand it before trying to assemble a bunch of words that communicates that understanding to others. Understanding the information is not a technology anyone has demonstrated to date.

LLMs, due to their ability to assemble grammatically coherent word streams based on their training data probabilities and the input quer(y/ies), are outright subversive in their simulation of intelligence. We all love a nicely constructed sentence and series of sentences. But there's no "there" there.

there is no known tech that could even remotely get that brain simulation into the same volume

Sure there is. It's called "biology." Nature has created brains / complex nervous systems over and over again, in quite a few different ways. We might end up doing so as well. Genetic understandings are developing and likely have a long way to go before we reach any limits (or implementations, but still, the tech potential is obvious.)

However, it's worth noting that there's no actual requirement for "getting [a brain] into the same volume." There's not even a requirement for having it cohabit the volume where its manipulators and sensors interact with the world as a robotic manifestation. We carry ours around, but current technology has already made that optional for robotics. RF is very handy for remote operations. A nice plus for this approach is if the body is damaged or destroyed, you just replace it and you still have the same entity.

The Guardian appears bereft of competent editors (thought... do the same people own them as own /. ? hmm...)

We all increasingly rely on smartphones, tablets, word processors, and apps that use autocorrect.

No, "we all" don't. That's one of the first things I turn off in system prefs of whatever device/software — computer, phone, pad, word processor, etc. I leave it off. Predictive suggestions are fine; that can be a timesaver. Altering what I typed... no thanks.

I'm not the only one on the planet's surface who sports better than rudimentary spelling skills and can hit the keys (virtual or otherwise) I aim at. Most of the time, anyway. Plus, there's that whole "read before you publish" process.

I'd subscribe to it (and I despise subscriptions, so...) if it would end the advertising on the various devices. I look over there to see what time it is and some completely irrelevant ad is up. Or, I just bought Only-Need-One-Of-Thing, now it's trying to sell me another identical Thing.

But my guess is "the beatings will continue until morale improves."

Someone pasted a sharepoint link into chat during a meeting yesterday, and it must have been 500 characters long. Sharepoint is a write-only system. The search is awful, and it's impossible to navigate by modifying the URLs, which is sometimes a shortcut for navigating other poorly-designed web-based systems.

Like so much that has been done in Windows in the last 10 years, this primarily serves Microsoft. Users be damned. There is nothing more sure to compromise users' security than a company like Microsoft promising that it won't compromise users' security. Given that they have been in quality free fall for some years now, there is no scenario where this Recall feature is not a disaster for users.

There is nothing that guarantees privacy will be violated more than a company like Microsoft promising they won't violate privacy.

Both these companies are racing to see who can hit rock-bottom in terms of quality first.

As the current LLMs are basically what we will have for the next few decades

I've enjoyed reading most of your comments on this post; very clear headed.

However, this bit seems to me to be unfounded speculation rather than assured fact. The velocity of machine learning development is very, very high. LLMs are also only one aspect of ML. We have already seen applications well outside of text prompting such as drug discovery and protein elucidation where ML has made significant advances. There are ML driven robotics being demonstrated right now that showcase the ability to learn in situ. It seems to me, at least, that the ML development curve is highly visible, monotonically upwards, and notably steep. Which leads me to think that there is very likely a significant impact to gainful employment in the relatively near-term offing. I just don't know how you could make the statement I quoted and be confident in its predictive ability.

If thing's[sic] are so crushingly expensive why is consumer spending at record levels right now?

When the expenses the consumer must disburse — food, fuel/transport, housing, medical care, education, tax collections, etc. — are subject to record level price increases for whatever reasons (corporate greed, increased/excessive governmental spending, usurious interest rates, transport costs, etc.) then the consumer will be spending at record levels to continue to do so.

Even if an individual is in the increasingly smaller portion of the population where these increased costs aren't a problem for them, they're still causing them to push more money through the system. Being wealthy doesn't make food less expensive, for example. So... record consumer spending. At every level.

Other metrics indicate that this record spending is not a healthy economic sign. For instance, rising credit card and auto loan delinquencies are signaling increasing distance between income and costs. People are spending everything they have to in order to get by and that naturally shows up as "increased consumer spending."

Another factor is people's inability to evaluate what is necessary. Large numbers of people treat various combinations of things like Netflix, premium sneakers and sunglasses, new clothing, high end phones, coffee, subscription software and services, visits to fast food emporia, bars and restaurants as "necessary." These are expenditures that don't help reduce spending on actual necessities in any way, but can in many circumstances cause the ready funds to come up short (and earlier) against those costs.

Social conditioning is largely responsible for these types of financial blunders, but again, given a previously somewhat stable situation that cost the consumer less, the increases will result in increased consumer spending until they hit the limits of what they can spend. They may then turn to credit, and we are seeing the results of precisely that in the current spate of increased credit delinquencies. Using credit to "get by" is unsustainable. But when people have to cover food and housing today, they will make the move today that ensures that is possible — today.

I find it incredibly disrespectful.

Then don't do it. Problem solved. You're welcome.

Some people question whether interacting with AI replicas of the dead is actually a healthy way to process grief

How about these assholes process their grief their way, and the rest of us will choose our own paths without them pretending to be our parents or guardians?

If your life consists of trying to figure out how to restrict the ways other people relate to their losses, your life is a net loss to society. Or, more succinctly, you're a shithead.