I think their point is that discounting the time estimates is more a constant shifting of the window of what we expect more than them being de-facto incorrect. They’re more off by degree (e.g. an XX% reduction vs complete extinction) than being worthless. As the example points out a large reduction can be very similar to an annihilation it’s just that we are only used to what we know so we constantly shift what is normal.
As someone actively researching in the neuroscience field these ideas are increasingly questionable. They do do a decent job of job of predicting neural data depending on your definition and if you compare them to hand built sets of features but we’re actually not even sure that will stay true. Especially in vision we already know that as models have scaled up they actually diverge more from humans and use quite different strategies. If you want them to act like humans or better reflect neural data you have to actively shape the training process to make that happen. There’s less we know about the language side of things currently though as that part of the field hasn’t yet really figured out exactly what they’re looking at yet because we generally know less about language in the brain vs vision. I think most vision scientists are on board with the idea that these things have really been diverging and have to be coerced to be useful. Language it’s more up in the air but there’s a growing wave of papers lately that seem to call the human LLM alignment idea into question. Personally I think the platonic representation idea is just a function of the convergence of training methods, data, and architectures all of these different labs are using. If you look at biological brains across species and even individuals within a species you see an incredible variety of strategies and representations that it seems ridiculous to me that anyone would suggest that there’s some base way to represent reality that is shared across everyone and every species. Here’s some articles that may be of interest if you’re curious:
If you could brain scan a human, and identify a shape of the network that corresponds to an emotion, and then could identify that in the ANN, could we say the ANN is experiencing an emotion.
I think its loosely referred to as "neural correlate".
I'm assuming what you are talking about with Convergence, would be these "neural correlate". And no reason we couldn't move beyond images to 'feelings'.
I think you’re treating the people at these companies as more dumb/powerless than they really are. I used to work in big tech and quit after a few years due to similar concerns that are being raised here. I will tell you anecdotally that everyone there I worked with thought our company was a net negative on society and that our work was at best indifferent and in my case likely exploited weak labor laws in poor countries to overwork people who we never were allowed to speak with for cheap data labelling. Yes, there definitely was some organizational shuffling to make it hard for us to see. We all knew, we weren’t idiots. My personal favorite book on similar concepts is “Modernity and the Holocaust”.
I would argue the organizational tricks exist more for the benefit of the worker than the org itself. The “powerless software engineers” there wanted the excuse to accept the huge salary for very easy work. The organizational tricks don’t fool anyone it’s a favor to the workers at these companies. They exist specifically to ease the cognitive dissonance just enough continue doing your job so you can get paid as much as you want without having to take guilt home with you. I’d say the same is true of my friends in the aerospace defense world. Do you really think they can’t put two and two together and understand that their “flight stabilization module” isn’t going to be used to blow up some school in another country?Your argument is just giving these people the ease of conscious which they want.
On the decision front as well I’d say most of the actual decisions in my org were not actually being made by C-Suite level or even executives. The managerial class at these companies are playing a totally different game than engineers. All the managers care about is that they have good metrics to show their boss so they can get promoted before the person on their sister team. I didn’t interact with a single person above VP and on my projects (as a recent grad mind you) I couldn’t even get my product manager to make a decision on how my product should be implemented. Everyone in the managerial class in these large companies largely exists to provide the illusion that you have no power. Meanwhile they have no idea what anyone in their org is actually working on and as long as they get a nice number to show their boss at the end of the quarter they won’t bother looking too closely.
I think we’re going to have a reckoning in the near future where we’re going to have to come to terms with the fact that the surveillance state which we’re scared of has been designed and built by the “powerless” engineers. The world is too complex for executives to actually have any understanding of what’s being done beneath them. There is SO MUCH room for the average engineer to shape their work in a more positive direction but that would actually require taking ownership over their work and risk some mental connection to its implementation. The average big tech employee already exists on the precipice of too much cognitive dissonance so they can’t afford to try and change anything otherwise they’d be convinced to give up their mid 6 figure salary while already having a larger net worth than 99% of the world will achieve in their lifetime. You cannot equate the life of a software engineer at a large company with the struggle of the working class in any meaningful way. Having a large mortgage is not at all similar to living paycheck to paycheck with variable hours at multiple jobs.
I’m being a bit brutal here I know but I’m so tired of people making excuses for themselves and others for living a life devoid of responsibility. If what I’m saying has struck a chord with anyone who is in a similar spot as I was, I’d suggest strongly questioning your position in the world. I have since found a different career path where I have clear ownership over my work and direction and am much happier now. I’m not fixing the world or anything and took a huge paycut but I agree with the outcome of job and am actually willing to work hard without resentment. I also applaud those who fight against the indifference of their coworkers in these companies since I know they exist. If every worker took responsibility for their output I promise you the parasitic Google or Meta as we know it would not exist. We are not the victims here. If we were desperate coal miners, I’d agree with you, but we are a class of workers with a level of financial flexibility, education, and freedom in our work most of the world has never been able to dream of. The success of these companies exists in how much responsibility they can put in the hands of their engineers who ultimately make most of the meaningful decisions whether they’re willing to admit it or not.
> I used to work in big tech and quit after a few years due to similar concerns that are being raised here.
This also reminded of a personal anecdote:
I had a colleague who was fresh out of college. He disagreed with whatever our company was doing at the time, but he said "at this point, fresh out of college, this ended up being the only company that hired me. I'm here for my CV only". The moment he could find a job elsewhere, he quit.
I think your point is well put, and I can agree with most of it, at least in terms of myself. I know I am a robotics engineer, and my best prospects for working on interesting or exciting stuff is in aerospace, defence etc. But I don't do that, because I think this way myself. I find it harder to judge others for the same.
That all being said, I think its worth noting that the reason that coal miners have historically had quite outsized political power and effect on politics was precisely because they required education, which helped them work together as a unit and demand better conditions and wages, as well as the camaraderie generated by experiencing bad conditions together. There are some great books about this, but the coal miners in the uk where im familiar with were much more educated than the average person, due to the engineering understanding involved. Yes it was bad conditions, but you can't have untrained workers using equations to figure out how deep into a rock face you need to put dynamite, and how much explosive of what kind, based on rock samples and tables from books. Same for what kind of supports and where, taking geological surveys etc. It was a high skill job, and also paid relatively well compared to manufacturing workers (not least due to said organization between miners)
The difference between then and now is that there is very little solidarity between software engineers. This is a state of affairs that I believe has been deliberately engineered. I do not think shaming people for where they work will improve things. I think one can assuage ones own guilt by choosing where to work, as you or I do. But I don't see how the solution can be to ask everyone in society to just not take a better life, better income etc for themselves. Especially when the harms are very indirect.
Politicians will do any draconian measure to help kids except try and improve the lives of their parents so that they can actually dedicate time to parenting. Making it slightly harder to access the internet fixes nothing. What if instead of having the largest prison population in the world our government supported communities that make raising good children possible? Our society needs to lose this urge to diagnose each other and provide some forceful treatment and instead set sights on providing the pre-conditions for everyone to prosper and lead their version of a fulfilling life. Only then will we have functional, healthy children. I quite like what the mayor of Baltimore has been doing to revitalize his city and it seems to be leading to actual change there if you want a good example: https://m.youtube.com/watch?v=XQs59YY-e2I&pp=ygUXY2hhbm5lbCA...
IF it were for the kids - but I don't think it is.
> Making it slightly harder to access the internet fixes nothing.
This assumes it is about the children. But if you do not think so
then it opens up new alternatives suddenly, be it from tracking
people, to targeted ads or any other information that could be
gathered and eventually either monetized or put in tandem with
other information. We'd get age graphs that way too.
Before that we could speculate to some extent, but with mandatory
age sniffing and id-showing at all times, those who track people
and benefit from it, benefit now even more.
> IF it were for the kids - but I don't think it is.
Exactly. If it was about kids, then you do not direct legislation towards everybody, but at parents. Give parents the software and tools (parental control settings) to restrict their child from searching certain terms and accessing certain sites. Bob or Susan (middle age adults), down the street, should have nothing to do with that.
It's about outright surveillance, tracking, censorship, control, and politicians putting money in their pocket. "Kids" is just a cover for their intentions and dirt. More has to be done to call out the deception.
Agreed. I’m sure some house members will vote for it because they only had a random staffer read the bill and heard that it gives them a good talking point in the next election. I just wanted to point out what’s maybe obvious to everyone that this won’t help kids. I’m sure this is being pushed by Meta/whatever other ad dependent business wants to pass off liability of verifying age with the added benefit to everyone in power that it’s easier to track everyone as a result.
The proving-your-age thing seems like a weasely way to talk about it. As you mentioned, providing a legible photo of your US state ID is a lot more data than your birth date!
well, and this bill literally only makes you prove age to ... set up the device.
how are we in 2026 and phones dont have guest mode or "i handed it to my kid mode"
apple's guided access is a terrible 1% solution to the problem. in one click i should be able to put my phone into some kind of locked down mode that exposes only what is allowed, starting with nothing unless whitelisted, with multiple profiles.
in the same sense, all the streaming services having their own separate kids profiles, instead of the streaming device having a single kids mode that exposes only the kids mode content from each app makes kids mode useless when a kid can just change the app, or gets stick into a single provider and i have to go help them switch.
In addition to profiles you can also 'pin' an app from the recents menu so the kid cannot exit the app, sortof like a kiosk mode. It requires bio/auth to undo the pin.
To me that's faster and much closer to a safe "hand the device to a kid" mode.
While age gate attempts are comically stupid, an adult giving a kid a device purpose built for addictive behavior should absolutely be as illegal as giving them alcohol or cigarettes. I really hope Apple and Google are not stupid enough to further enable this.
> Politicians will do any draconian measure to help kids except try and improve the lives of their parents so that they can actually dedicate time to parenting.
Because in their eyes your children are not your children. You are simply a custodian of their future work force asset. If you educate your children too much into individualism, they (today’s politicians) may see a diminished return of whatever they want to achieve.
And if you don’t agree with me on an emotional level, well, just remember the words of Elon Musk (paraphrasing): we need people to have children because we need to have workforce in the future. Translation: we need people to have children because who will work for us and makes tons of money.
If you have it too good, you aren’t dependent on them, you have all the carrots. They have no stick. They want to have the stick.
Sure, but this is only tenable as a technical position that aims to reduce all forms of centralized power. It completely fails as a political position applied to the nominal "government". Politically, pushing in this direction seems to only ever play out as reducing the power of governments over corporations, while often even increasing the power of government over individuals (spurred on by corporations looking to wield that power through the government). Whereas for it to achieve its intended individual liberty, the complete opposite would have to happen - decreasing the power of governments over individuals while holding or even increasing the power of governments over corporations - otherwise unrestrained corporations simply step into that nonconsensual role of government and we're back to step #1.
Would you believe that I agree with what you wrote completely?
I’m pretty much a pure anarchist in terms of principles, but I’m a pragmatist in practice. I’d describe my approach in politics as “What do you wish the government would stop doing? Let’s focus on making that happen.”
You can’t change a culture by changing the political system, but my hope is that you can change a political system by changing the culture. I want to be as independent of the state as I can possibly be, and I want to encourage others to do the same. My hope is that this sort of cultural shift will eventually lead the shrinking of the state. I don’t expect to live to see that happen, but I hope my children and their children do.
Aside from the above, just don’t harm others. That’s it.
I would. My own views had to come from somewhere, right?
Responding to what you've said, my unfortunate experience is that culture always ends up going sideways. As movements grow in mindshare they tend to attract people focused on power/expedience, only applying the initial precepts towards those ends. And gaining control over some existing centralized power structures is much more lucrative than a given person's share of the distributed wealth that would be created by successfully constraining them.
Which ties right into the problem I saw with your original comment. A statement like "Government should be so powerless as to be unattractive targets for corporate influence" lands in the political/partisan context by default. And while perhaps that's a symptom of how [unfortunately] inured in centralized politics we are, it's still a fact. So even though we can both take a step back and lay out the context where that can be an agreeable productive statement, the overwhelming use of similar statements is actually to attack individual liberty by getting people to overfocus on the nominal government while giving a pass to another primary contingent of the centralized power structure.
If you’re interested, see my response to a sibling comment of yours for a more complete description of my mindset.
Specific to this, though: there’s a big difference between a stateless society and a failed state. You’re describing failed states.
I also very much agree with you about the result of a power vacuum. I argue that a power vacuums exist not because of the absence of a state, but because of the absence of a state where the populace expects and relies on a state to be.
I didn’t say we should get rid of it all tomorrow morning :)
> Politicians will do any draconian measure to help kids except [...]
They are covering for and not prosecuting perpetrators in the biggest child trafficking and abuse scandal in recent memory -- the Epstein case. Let us do away with even a surface-level pretense that they care about kids at all.
There’s a lot of research out there about the general flexibility of the brain to adapt to whatever stimulus you pump into it. For example taxi cab drivers have larger areas in their hippocampus dedicated to place cells relative to the general population [1]. There’s also all kinds of work studying general flexibility of the brain in response to novel stimulus like the visual cortex of blind people being dedicated to auditory processing [2 is a broad review]. I guess you could argue that the ability to be flexible is intelligence but the timescales over which a brain functionally changes is longer than a general day to day flexibility. Maybe some brains come into an initial state that’s more predisposed to the set of properties that we deem as “intelligence” but development is so stimulus dependent that I think this definition of a fixed intelligence is functionally meaningless. There are definitely differences in what you can learn as you age but anyone stating we have any causal measure of innate intelligence is claiming far more than we actually have evidence for. We have far more evidence to suggest that we can train at least the appearance and usage of “intelligence”. After all no one is born capable of formal logical reasoning and it must be taught [3,4 kind of weak citations foe this claim but there’s a lot to suggest this that I don’t feel like digging up]
Agreed. I find that people who argue that religion is necessary for ethics tend to ignore the history of their religion and the fact that the original text largely serves as a jumping off point for religious philosophers to connect older “secular” texts to this new religion. Modern Christianity is a complex combination of Platonic, Aristotelian, Syrian, and Roman ideals which are taken out of their original context to align with the Bible even though the original writers would say they knew nothing about Jesus. The base texts which many of these ideas are based on make almost no appeals to God and focus more on what it means to live a “good life”. To be fair a lot of great ethical arguments are made by Christian writers but I think that’s more just a consequence of their cultural upbringing and the fact that the thing the New Testament really added to the discussion was that your ethical responsibilities generalize beyond yourself and your friends/family.
Religious ethics are just as fluid and complex as secular ethics, it’s just that the concept of God makes people think they can claim that their way of thinking is the only one that’s real. I would guess if you self-reflect though you’d see that even within one lifetime the definition of what’s moral in a religious context changes as well.
This is an interesting insight I hadn’t thought much about before. Reminds me a bit of some of the mechanistic interpretability work that looked at branch specialization in CNNs and found that architectures which had built in branches tended to have those branches specialize in a way that was consistent across multiple training runs [1]. Maybe the multi-headed and branching nature of transformers adds and inductive bias that is useful for stable training over larger scales.
This is a large claim to make without any evidence that is eerily reminiscent of historical arguments in favor of eugenics programs. Society thrives off of diversity and variations in thinking patterns. Dividing all people into either neurodivergent vs neurotypical or in your preferred terms “brain-damaged” vs “non-brain-damaged”is a vast oversimplification of the reality. Your claim about neurotypical people taking over society in the 20th century isn’t supported by evidence that suggests that genetic markers in humans for things like Autism Spectrum Condition are downsampled in humans relative to other species suggesting that human evolution has selected for some of the traits of what we classify as neurodivergence while balancing out the effects of some of those traits [1]. I’m not trying to say there’s anything wrong with neurodivergence (that’s how I’ve been classified) but this dichotomy is dumb. Everybody is neuro-divergent and what we define as neurotypical is societally defined. You’re just trying to flip the idea of what’s assumed to be a “good” person and that need to declare one group as better than the other is the actual problem. Please read more about the wider variability in cognition among humans before claiming that anyone in your preferred definition of “neuro-divergence” is superior to others.
I'm not talking about genetics. There must be something that causes brain damage (and it would probably be a good idea to find it and stop it). The change was too fast for genetics. It was essentially one generstion mostly normal, the next one brain damaged.
I'm talking about all the mid 20th century talk about the "generation gap", "teenage rebellion", and so on. Essentially no stone was left untouched in how the society worked between the "conformist" 50s, and the 70s. Something happened then, or a bit earlier, and it's still doing the damage.
The book Modernity and the Holocaust is a very approachable book summarizing how the action of the holocaust was organized under similar assumptions and makes the argument that we’ve since organized most of our society around this principle because it’s efficient. We’re not committing the holocaust atm as far as I know but how difficult would it be for a malicious group of executives of a large company quietly directing a branch of 1000’s who sleepwalk through work everyday to do something egregious?
reply