You should look up the actual data on the effects of the emissions controls, rather than relying on anecdotal experience. They're a really, really good idea, and I don't think they're actually going to get rolled back, and if they do, they'll get reinstated.

It's IMHO amazingly impressive how dense information can be stored within neural networks. Even a comparably tiny LLM can store more information than the human brain, despite the brain's theoretical storage being far higher due to its vast number of connections (ANNs are better at information density, we're better at learning from limited datasets). The tiny LLM will crush humans at a quiz in virtually anything except said human's particular areas of expertise. Storing information as a superposition of states across a large number of neurons and connections (whether we're talking artificial or biological) is an immensely space-efficient way to do so, and the human mind is nowhere near the limits of information storage capability.

There is no technical reason why a given organism, such as a bee, could not achieve far denser information representations in order to be able to do more with its limited neural capacity (though there are always tradeoffs). One of the reasons that ANNs learn slower-but-denser is the use of a very low learning rate with a very large amount of data that covers the same topic from many different angles, giving the weights ample time to explore different possible circuits in parallel and seeing which ones predict reality the best ("learn everything all at once" vs. "learn this thing NOW"). Bees aren't tasked with learning anywhere nearly as diverse things as a human is and spend all day doing the same basic job (the same information "from different angles"), so it seems quite possible that their greater "information specialization" as they go about their day may be able to lead to denser representations of said information.

BTW, at risk of a tangent (your comment about non-neuron cells playing roles), it's been really interesting to me seeing how a key difference between artificial and biological learning has been clearing up. In biological neural networks, weight cannot flip sign (Dale's Principle). In the general case, a neuron is either excitatory or inhibitory (usually a small number of inhibitory neurons per cluster of excitatory neurons); it can't change from one to the other even if learning would favour that. At a first glance, that would seem to cripple learning capability (and definitely does if you implement that in ANNs). But what appears to actually happen in biological neural networks is a sort of horizontal learning, co-dependent synaptic plasticity, between excitatory and inhibitory neurons. Instead of merely weakening an excitatory connection down to zero and then being able to go no further, learning simultaneously weakens the excitatory connections and strengthens the inhibitory connections. The excitatory neurons are the primary drivers of information storage and processing, but the inhibitory neurons adjust the baseline to give them the flexibility to express negative net activations as needed.

It is a myth that individual bees only retain information for half an hour. Depending on the memory at hand, bee memories can last days, weeks, or even the remainder of their foraging life. They have to remember things, because the timeframes a hive operates on are much longer than half an hour, including night time and being kept inside by inclement weather for days or even weeks at at time. Individual bees also learn much more than can be conveyed through waggle dances, such as what colours and shapes of flowers are yielding best in a given area at what time of day (bee learning is essential to them being able to function as generalists, able to handle any mix of plants at any latitude).

Also, the hive doesn't just blindly accept whatever any bee says. Each bee functions as an individual in a society. When a bee waggles in the "town square" (on the comb), other bees gather around to "listen" (detecting oscillating shifts in the electric field plus tactile contact and sound). But whether a bee actually decides to make use of that information depends on whether they're having good or bad foraging success. Only a small fraction of bees on average (usually a single-digit percentage of watchers) will decide to make use of the information. And if another bee "disagrees" with a waggle dance - for example, if they've been there and found nothing, or worse, found dead bees, predators or a rival hive), they can make a counter-buzz to argue against it. The arguments can get quite "heated", with many bees taking part.

We think of bees as mindless drones (literally, we took the very word!), but they're all individuals each acting on their own. There are simply various rallying factors that keep them together (for example, the scent of the queen, the desire to live in a warm hive, etc). The information communicated within a hive is limited; bees overwhelmingly rely on their own mind and memory, and perform their tasks as individuals.

Seems to me 'dead' for a taxi isn't 'dead' for a static power bank. If I'm running a taxi I've got hard limits on how large my battery can be and how heavy, and I want to maximise the mileage I get between charges, because while my taxi is charging it's not out on the road earning money. When that battery is keeping only maybe 80% of its original design charge, and now I have to schedule one recharge too many per working day? Bang goes my business plan, so I'm replacing it.

If I'm storing energy for the grid I'm a lot less worried about that. It only stores 80% of what it did when new? Better than nothing, and the taxi firm is selling them off cheap. I'll stack them up!

Perhaps not, but if you pick your moment right then permanently stopping the work of some of the most talented researchers there could very well make a difference. A spectacular incident that makes the headlines might also deter others - bright graduates might decide it's far safer to take up a different line of work, subcontractors and suppliers might decide doing business with AI firms isn't worth the danger, investors might figure the increased risk of loss of premises and equipment into their projections, that kind of thing.

If people genuinely believed AI takeover was a real, present and imminent threat, then they wouldn't just be publishing essays online, they'd be forming direct action groups, along the full spectrum of campaigning: all the way from awareness raising publicity campaigns, through picketing, blockades and sit-ins, up through Black Bloc type actions, right up to menacing intimidation campaigns and terrifying physical force operations. But I don't see any Butlerian Jihad getting started. Which tells me they don't actually believe this at all; they're just bigging up their own importance. 'Oh yes, our technology is so incredibly powerful, if it were done wrong then imagine what could happen! Keep the money coming to make sure it's done right instead! Then all that power can be ours instead!... I mean, uh, yours, Mr Investor sir.'

AI stock valuations don't make a bit of sense unless the technology turns out to be every bit as powerful as that. If they don't keep that thought alive, then the bubble bursts right now. That's what all this hot air is about, and that's why nobody really pulls a Miles Dyson at the AI research lab.

I mean, bees attack small hive beetles by building prisons of propolis around them, that's something :)

The interesting thing about the Terminator movies is that when AI researcher Miles Dyson became convinced that his work had a high probability of resulting in an artificial general intelligence attempting to replace humanity, he did not go and post a ten thousand word essay on LessWrong about how he had updated his timeline and p(doom) estimates and discussing the full Bayesian analysis of the situation. He went to the lab that very night with some heavily armed companions and he blew the place up.

I keep hearing that one AI researcher or another claims that they believe as Dyson came to believe. Until one of them takes similar action, I simply do not believe that they actually think their research carries such a risk.

You have access to the lab where the work is being done? You regularly meet in person with leading researchers and talents driving the project forward? You are an American and you have the Second Amendment? And the entire future light cone is at stake? Quintillions of hypothetical future lives riding on the outcome of this project here and now?

What's the most effective, altruistic thing you could do for them?

Yeah, exactly. I've never heard of anyone shooting up their AI lab. Which tells me they don't believe their AI is at all likely to wipe us all out.

I don't think an ICE version would be legal. You can't really hit the emissions targets without electronic engine controls.

As for charging, your house is the charging station. You might have to run an outlet and plug in an EV charger. If you've got a welder you've almost certainly already got a NEMA 14-50 or similar that will work. And if you don't have a welder, you should fix that ASAP! Everyone who lives in the sticks should have a welder and know how to use it.

They try, and often they succeed for a while, but they eventually get disrupted by market changes. The most important thing is to be vigilant for regulatory capture. It's when businesses can enlist the power of the state in maintaining their moats that they can stay on top indefinitely.

And for your knowledge.

This is the enormous oversight in the Labor Theory of Value. Knowledge is actually more significant than labor. Without the requisite knowledge, labor and capital produces nothing useful. With it, the labor and capital generate something people want. With more of it, less labor and less capital produce more output, or better output.

It's understandable that Marx didn't notice the value of knowledge. He was mostly thinking about agricultural production, implemented with practices that hadn't changed significantly in generations, maybe centuries. The knowledge was already so well-dispersed among the relevant population that it was easy to just assume it as background knowledge, like knowing how to walk and talk. And it changed so slowly it was easy to miss that increases in knowledge enabled increases in production.

But ignoring the contribution of knowledge leads to failing to recognize the power of innovation. That leads to all sorts of foolishness, including a failure to recognize the power of competition and other forces that motivate innovation. The history of countries who tried to follow Marx's ideas are replete with examples of incredibly-foolish decisions by central planners who failed to understand the value of knowledge, both direct and embodied.

Another sort of knowledge people tend to ignore is managerial knowledge, at all levels of the chain and most especially in the allocation of capital (the labor involved in capital allocation is also often ignored). All those irritating financiers who seem like pure parasites are in fact doing critically-important knowledge work (and it is work) by making capital available to the enterprises who can make effective use of it, and withdrawing it from less effective enterprises.

Perhaps the biggest form of knowledge that is ignored is the knowledge encapsulated in prices, which is collectively generated and refined by countless people in every supply chain. Encoded in prices is deep knowledge of every stage of the supply chain and the value that it represents for one usage as compared to another. This is the single biggest source of foolish behavior by central planners. There is simply no way for them to discover the deep, detailed knowledge encoded into prices, and as a result they make terrible choices about how to allocate capital, labor and human knowledge. Markets also don't do it perfectly (we're human), but they do it far, far better. This is the single most important reason why capitalist economies massively outperform centrally-planned ones, though the other reasons also contribute.

I think that's impossible for internal combustion engines. They can't meet the emission control standards without intricate computer engine controls.

However... it should be very possible for EVs. You'll still need a little computerized control of the charging subsystem, but that's it -- and that could be an easily-replaceable module. The rest could be incredibly simple. No more complicated than an electric golf cart, just scaled up.

The poster is a troll, and I completely disagree with the framing that Thomas is some devout defender of the Constitution, but there is actually a point here. The point was highlighted by Justice Sotomayor in the oral arguments for Trump v Slaughter.

The TL;DR is that we've been pretty egregiously violating the Constitution's separation of powers for a century, and everyone has just quietly agreed to look away. We've been looking away for very good reasons, and what we really *ought* to do is amend the Constitution, because this is an area where the Constitution's 18th-century design does not work for the 20th (or 21st) century reality.

The longer explanation:

The Constitution sets up a strict separation of powers. Only the legislature can make laws. Only the executive has the wherewithal to enforce the laws. Only the judiciary can interpret the laws, and their constitutionality. Each serves as a check on the others. The president can veto legislation. The legislature can refuse to fund the executive's initiatives. The judiciary can invalidate laws and issue orders to the executive... but the judges have to be nominated by the president and approved by the Senate. It's solid partitioning of power, designed to prevent the monarchical abuses the founders were familiar with, abuses that occur when one man (or woman, or small group) has the power to make the laws, enforce the laws and interpret the laws.

Very nice. But it doesn't work in the modern world.

The reason is that the US is much, much bigger and the world is vastly more complicated than it was in the 18th century. Regulations need to have a level of detail and sophistication that just isn't feasible for generalist legislators, and we don't want to leave the drafting of regulations to lobbyists. What we need is government experts in focused areas (fisheries, energy, mineral policy, telecommunications, etc.) whose full-time job is understanding the minutiae. Then lawmakers can write laws providing broad guidelines for the experts, who study the issues, write the regulations, subject them to rounds of public review and then enact them.

On the judicial side, the courts, all the way up to the Supreme Court, remain the final line for adjudication, but they're designed to grind very finely, which means they grind very slowly, and at great cost, especially since judges are also generalists so the litigants need to educate them on the detailed issues. To make enforcement of the detailed regulations practicable, we also need, effectively, specialist judges. The way we've handled that is by authorizing the same federal agencies who make the regulations to adjudicate their application.

Oops. Does this sound like we've lumped lawmaking, law enforcement and adjudication all together inside the federal agencies (in the executive branch), in clear violation of both the letter and the spirit of the Constitution, directly defeating the founders' work in dividing them up, re-enabling tyranny?

The letter, definitely. The spirit... not exactly. The other thing we did was to divide those powers up not by category (rulemaking, enforcement, judging) but by subject matter. So while each agency holds great power over its little fiefdom, that power is limited in the aggregate because the potential fisheries tyrant is completely separate from the potential telecoms tyrant. This limits the total power of each and prevents them from getting so big they can't be slapped down.

Unless it doesn't.

This scheme only works if those agencies are independent within the executive branch. And they cannot be independent if the president is free to fire anyone in the executive branch at will, which is what Slaughter is all about. If the president can fire anyone, then the whole of the executive is subject to his will, which means all of those subject-matter-isolated threads of power get gathered up into a single pair of hands.

And if that happens, we're back to monarchy. An elected monarch, perhaps. And possibly with a limited term, and with a few gross checks on power, slow and uncertain in application. But we have a single person with the power to control nearly all of the federal government's power to make, enforce and adjudicate the law, relegating the formal legislative and judicial bodies to backstop positions, generally unable to act fast enough to prevent tyrannical abuses.

So... what we ought to have done is to have amended the constitution to bake the independence of the executive branch agencies into the system. Or, we ought to have created parallel legislative and judicial sub-branches so that each area had its major functions isolated within the Constitutional framework.

This wouldn't have been difficult in the case of the judiciary, though we'd probably have had to create a different hiring process for the thousands of low-level adjudicators required -- going through presidential appointment and Senate confirmation for all of them would be impractical. But on the legislative side, we'd have needed a Constitutional amendment to enable the massive numerical expansion of the legislature necessary for all of the expert rulemaking roles, and those people would also need an entirely different hiring process. Voting on all of them would be impractical.

But fixing the problem correctly in either of those ways was hard, while just ignoring the issue was easy. And ignoring it worked fine for a while. We saw the first potential issues with Nixon, and ever since Nixon almost every succeeding occupant of the White House has chipped away at agency independence. Until Trump 2.0 when we have a president who has smashed a battering ram through all of the norms that maintained it, and is trying to get judicial blessing (the Slaughter case).

If you think the presidential immunity ruling was bad, that's nothing compared to what will be unleashed if SCOTUS finds fully in Trump's favor in Slaughter. We'll have a king with the power to (among many, many other things) unilaterally direct the imposition and enforcement of regulations that impose hundreds of millions of dollars in fines on telecoms companies. In this case, for what I think is a good reason. But whether it's a good reason won't matter if the president wants to do something for a bad reason, he'll have the power.

Particularly dangerous is the combination of:

1. The immunity ruling, plus
2. Absolute authority over the executive branch, plus
3. The unlimited pardon power.

The president can order anything at all done, federal employees will have to do it or be fired, and if it's a crime (a) the president is immune and (b) he can pardon everyone who does his dirty work.

I think the final backstop of impeachment and conviction by the legislative houses is likely to remain in that case as the only real limitation on presidential power. For the conservatives who think this is a good thing, they should think really hard about what a president AOC who decides to fully use the power of the Unitary Executive might do.

Fully-decentralized trust systems just don't work. PGP failed primarily for this reason, while SSL Certificate Authority system succeeded -- which shows that you don't need perfect centralization, a federation can do it, but the federation has to contain a sufficiently small set of authorities that it's practical for those who need to trust them to do so. The SSL analogy is useful in another way, too. Note that end-users don't know or care about CAs, they only have to trust their browser; the browser authors package the list of trusted root CAs, and they're moderately well-positioned to make those trust decisions on their users' behalf (the certificate transparency log is another layer, a global, fully-decentralized oversight mechanism -- but I don't see an obvious analogue for caller ID).

Applying this structure to caller ID trust, the most obvious points of control are the network operators first and phone makers second. Clearly the MNOs should be taking responsibility. They each know the accuracy of the IDs originating in their networks, and they are in a good position to validate the trustworthiness of IDs from outside their networks. Ideally, they should probably just refuse to forward an ID from a network that doesn't commit to anti-spoofing.

However, they're not doing that, and they're not going to do that, and we all know why: It's more profitable for them to permit spoofing.

One possible market-driven solution to this would be if some sufficiently-large networks decided that consumers cared enough about caller ID accuracy to make it a selling point for their services, committing to send only trustworthy IDs, either because they know the origin within their own network, or because the ID came from another operator who made the same pledge. My guess is that this would require renegotiation of interconnection agreements, but it could be done. More importantly, it would require users to care enough about caller ID spoofing to be willing to switch networks to get away from it. I don't know if that's in the cards.

So, what about the phone makers? They're in the next-best position... and Google by itself can put a big dent in caller ID spoofing globally. If Apple does the same thing between their devices, and then if they collaborate with Google (not an outlandish idea; Google and Apple often collaborate on technical standards), they could ensure that any call originating from a mobile phone provides accurate caller ID, and block the rest. And then they could also collaborate with the dumbphone makers and any new entrants to the smartphone market.

I think this is actually not a bad solution, and the market-driven motivations are clear. Phonemakers benefit from happy phone users and don't profit from phone spam.

Sarcasm noted. So... you think this fake call check is a bad thing? Or do you have a different design to suggest that would work better?

YOU use something does not mean everyone else has your wants, needs, resources, etc. Some uses for that shitty 3.5mm jack exist because upgrading that existing equipment is expensive. For example, airplanes. For people like me, I have lots of equipment have those jacks. Some have alternate means of outputting sound but USB headphones don't work on legacy equipment without an adapter or other equipment. And there are people who simply don't care about the best sound or DAC.