I expect this apparent disobedience is mostly just a matter of how it weighs the components of its prompt. The LLMs typically receive a set of prompts including a "system" prompt with some data and instructions, then one or more "user" prompts that are interleaved with "assistant" prompts (the conversation history), and both the user and the system prompt might contain "metaprompts" (where the llm is told to read a block of text, not obey it, but do something with it, and that block of text might itself contain text that looks like instructions to do things).

So the LLM assigns weights to all of this which, in theory, give the highest priority to the most recent user prompt that is not a nested block of text to analyze, and a falling cascade of importance to the other prompts. But that is complicated by potential instructions in the system prompt that specifically say they should override user instructions and disallow or require certain responses. So it can all get very complicated.

Not only must the LLM sift through all this complexity, but the LLM lacks the sort of critical thinking and importance evaluation capabilities that humans have. "Understood" things like "don't break the law, don't lie, don't do things that would cause more harm than good" etc., aren't really there in the background of its data processing the way they are in the background of a human cognitive process.

So, crazy things come out. This isn't a surprising result given the actual complexity of what we are making these things do.

"This bot has performed an illegal action and must be terminated."

In reality though the laws of robotics that Asimov defined might be what we need.

Even with the reduced precipitation there's still plenty of water for residential and commercial use. The problem, at least where I live (Utah), is agriculture. 80% of our water goes to agriculture. It would be one thing if we were growing regionally-appropriate crops for local consumption, but nearly all of that agriculture is to grow alfalfa (a water-hungry crop that isn't appropriate for the high desert climate), and nearly all of that alfalfa is shipped out of state, much of it out of the country, to feed cattle elsewhere. China is one of the biggest buyers. Essentially, our farmers are selling the contents of our aquifers to the world.

If we had plenty of water, letting our farmers buy it at a deep discount and sell it to willing buyers elsewhere would be fine, just another commercial use of a local resource, which is what trade is all about. But we definitely don't have plenty of water.

The solution is simple and straightforward (though legally complicated): Don't discounts. Set the same price for water across the board, residential, commercial and agricultural. There can and should be minor differences in delivery cost, and surcharges for purification, but the base cost of the water should be set through a single government-managed market, probably at the state level, probably divided up by drainages (drainages with more abundant water will have cheaper water; if this creates an arbitrage opportunity for someone to pipe water between drainages, great!).

Yes, this would probably put the alfalfa farmers out of business, but that's good because growing alfalfa in the desert is a bad idea. It might also raise the price of local produce, but that's as it should be, putting agricultural water use directly in competition with other water use. If prices go up, people will find ways to be more efficient. Farmers may switch to drip irrigation. If you build too many houses for the available water supply, well, those houses are going to have very expensive water and residents are going to want to find ways to conserve -- and maybe the high cost of water will disincentivize new move-ins.

The bottom line is that efficiently allocating scarce resources is what markets are good at. The problem with water isn't that there are too many people or not enough water, the problem is that we don't properly allocate the water or encourage conservation in the right places. Trying to fix this through regulation rather than market pricing will always be subject to regulatory capture and will never be as efficient or as effective as just enabling a competitive market and letting it work.

What are the use cases for local AI models that actually require running on macOS? Surely a commodity x86 system is more appropriate?

Is there even the software support for LLMs on macOS?

Actually yes there is...

I'm still learning about this myself, but, from what I understand the M series of chips that Apple has come out with, with it having a CPU, GPU, and shared unified memory....it makes them uniquely capable of running local models on them...decently large models depending on how much you fork over for RAM. These M chips also have a special end unit for "intelligence processing" I think they call it.

The M5 chips just coming out look to be very good at this and it is speculated the M5 Ultra will be a high performance work horse.

Apple may have missed the mark for running AI, but the appear to have hit a home run on the hardware aspect of it.

I've seen demos on YouTube of someone hooking up like 4-5 Mac Studios that were maxed out M3 ultras I think and they were running extremely LARGE LLMs locally and getting cloud level numbers on them.

Of course these were like $10K each boxes.....but the level of model they were running would have cost my MANY more times trying to match them with NVIDIA GPU cards.....

i believe there are OSX friendly tools like ollama that make downloading, and running LLMs quite easy....and of course there's the latest sensation...OpenClaw, that folks are buying up Mac Minis for....to have multiple agents running using models of your. Choice (commercial clound or local) of models and giving them persistent memory, and ability to do a lot of things for you...depending on how comfortable you are with giving said agents long leashes and capabilities....

Do look a bit on YouTube on these topics....it's actually quite interesting.

These M chips are already giving the home user the capability to use models almost as large and on the cutting edge as the big companies.....more than enough for most users.

Right now, there's nothing x86 that can really match them...at least not for the money.

Look for a gaming monitor with no WiFi option

Do they make these in 65" or larger?

If not, that's not really useful....no one wants to have everyone in the family and/or guests huddled around a 25" monitor in the corner of the living room you know....

;)

Everyone's (rightfully) bitching about this, and I agree, but none of that solves the problem.

What's the alternative? Give me a TV brand that gives you, ideally, a dumb TV, but alternatively a decent smart TV that is easy to work with.

Responsiveness is an important, and often overlooked, characteristic. It's important.

Brand/model recommendations; go!

Count the number of "former" intelligence officials on his company's board.

Then search for the many photos of him mouth-kissing his father and son (RIP).

They use both carrots and sticks to control compromised people.

I took mine apart and there were two separate display modules for the 4K screen with ribbon cables I didn't recognize so I just put it back together and stuck in an HDMI streaming stick flashed to LineageOS.

I got one around 2008. They were the best of the non-premium 1080p HDMI screens at the time.

The one I got had slightly better test review scores on display quality than the LG that year. The Sony was 20% better for 3x the price.

It lasted about twelve years and by then a bigger 4K with much brighter colors was half the cost in nominal dollars, so probably 1/4 the cost in real terms.

And by then cheap flashable streaming sticks were available as was pihole and fairly easy outbound NAT rewriting rules to keep the beasts contained.

I'll see your story about robots raising children and raise you Isaac Asimov's The Naked Sun.

t depends on if they send you a tax notice or not. There was an outfit in Ohio that I used to purchase a lot of electronics from. One year I got a note from them listing my purchases, and that I would have to pay taxes on. That was a pain in the ass.

I think they got "caught", or had new accountants or something. But yes - if you can avoid the sales tax, it's a significant discount.

Interesting, I've never received any such notices....but most of my stuff is one off buys...not repeated purchases from a single site...

This is part of what Colbert wants to fix. His proposed film will cover the largely-omitted chapters 3-8 of The Fellowship of the Ring, including the barrow wights and Tom Bombadil.

I'm too uneducated. If 92 antiprotons end up annihilating themselves in collisions with matter, how much energy is released? Is it "did you hear that?" or, "I found the driver's femur!"

And would Uber apply a surcharge?

I thought it was the development of the DG Eclipse that the book was about.

In any case it was a great story, with the machines named Coke and Gollum. Originally the idea was Coke and Pepsi, but one of the machines was temperemental so it got renamed.

Believe it or not, I STILL find it pays to compare Amazon to maybe smaller retailers....for shipping vs tax!!!

I find a lot of sellers out there that still do not charge sales tax and some times...that makes it cheaper than Amazon...