So the two biggest examples I am currently aware of are googles AI for unfolding proteins and a startup using one to optimize rocket engine geometry but AI models in general can be highly efficient when focussed on niche tasks. As far as I understand it they’re still very similar in underlying function to LLMs but the approach is far less scattershot which makes them exponentially more efficient.

A good way to think of it is even the earliest versions of chat GPT or the simplest local models are all equally good at actually talking but language has a ton of secondary requirements like understanding context and remembering things and the fact that not every gramatically valid bannana is always a useful one. So an LLM has to actually be a TON of things at once while an AI designed for a specific technical task only has to be good at that one thing.

Extension: The problem is our models are not good at talking to eachother because they don’t ‘think’ they just optimize an output using an intput and a set of rules, so they don’t have any common rules or internal framework. So we can’t say take an efficient rocket engine making AI and plug it into an efficient basic chatbot and have that chatbot be able to talk knowledgably about rockets, instead we have to try and make the chatbot memorise a ton about rockets (and everything else) which it was never initially designed to do which leads to immense bloat.

Yes, my whole post was that non-LLMs take far less processing power.

The crazy part is outside LLMs the other (actually useful) AI does not need that much processing power, more than you or I use sure but nothing that would have justified gigantic data centers. The current hardware situation is like if the automobile first got invented and a group of companies decided to invest in huge mortal engines style mega-vehicles.

So you’re getting a lot of downvotes and I want to try and give an informative answer.

Its worth noting that a most (it not all) of the people talking about AI being super close to exponential improvement and takeover are people who own or work for companies heavily invested in AI. There’s talk/examples of AI lying or hiding its capabilities or being willing to murder a human to acheive a goal after promising not to. These are not examples of deceit these are simply showcasing that an LLM has no understanding of what words mean or even are, to it they are just tokens to be processed and the words ‘I promise’ hold exactly the same level of importance as ‘Llama dandruff’

I also don’t want to disparage the field as a whole, there are some truly incredible expert systems which are basically small specialized models using a much less shotgun approach to learning compared to LLMs that can achieve some truly incredible things with performance requirements you could even run on home hardware. These systems are absoloutely already changing the world but since they’re all very narrowly focussed and industry/scientific-field specific they don’t grab headlines likes LLMs do.

Depends how many European politicans are willing to deliberately sabotage their own countries defence in exchange for personal wealth. And I know I’ve made several comments like this which could be interpreted as trying to sow disent but having seen what my own government gets up to I’m genuinely worried those assholes will sell us all into gulags for pennies a head.

Not american but if ‘lets not aid genocide’ is a magic pony level ask then you guys are already fucked beyond help.

Except that every time we print money for some bailout or crooked public works project that never materializes we further devalue the currency and all that money goes to the rich. Simply by being so insanely rich they’re already causing a decrease in everyone elses wealth.

The alternate take I’ve heard is to take Venezuela out of the oil market entierly thus driving up the price and propping up Russias economy for a bit longer.

Iirc this is an actual (as in offical and described in legislation) lever the EU can pull, unfortunately regardless of public sentiment many of its leaders are beholden US economic interests.

On the air defence note, specialist systems are already amazing at tasks like this and can usually run fine on high end consumer tier hardware (up to a single server rack) rather than data centres with the power requirements of small countries. AI is here to stay but LLMs are absoloutely a bubble.

China is just state capitalism.

If it comes to war with Russia I half expect my country to just immediately surrender in exchange for a personal payment of 5 whole rubles to each member of whichever political party is in charge at the time.

We can finally make endo steel chasis!

My dream phone would basically be a phone jammed into a small handheld console with a big battery. Like stuff a pixel into a gameboy advance and that’d be perfect.

Almost like we’ve already pretty much optimized the design and trying to fix what isn’t broken just leads to stupid consiquences?

This is why I only generate text using a randomiser based on the detection of energy released from virtual particle pair annihilations.

to be fair on traffic engineers timing lights we have verifiable proof that sometimes adding roads can actually increase taffic (and that removing them can decrease it) And this information is not always taken into account on a high level so they get stuck trying to fix an innately flawed system.

So if anti-genocide is ill anti-legal then the logical conclusion the government wants us to draw is?

I’d agree with that when it works. When linux setup works its great, when it doesn’t work getting it working again is obscure as hell, Windows almost always sets up correctly first time but its obscure as hell to not make it be kind of shitty.

I’ve said this before and I’ll say it again. What Linux needs is a straight forward setup. Yes Mint is normally super easy to install but can also randomly just not work due to what is often a very simple issue but one obscure enough that the inexperienced (like me) will take hours or even days of trying different solutions until they find it. I love how light linux is but an extra half a gigabyte in the setup to just innately include solutions to the most common issues would pull in way more people than it would push away.