This happened after I stopped working for the DOD, but I am 99% sure the program in question during this incident was something I worked on. It was effectively an anomaly detection system for identifying any weird behaviors detectable in WAMI (wide area motion imagery, which is high resolution, low framerate drone video data that covers a large geographic area). As an example from the unclassified sample project they asked us to do as part of the bidding process, our code flagged a car doing donuts in a parking lot (the sample data was just from a random US city on a random day because this was an unclassified open bidding process). The system was not and never was intended to be a "terrorist" detection system. The idea was always to flag potentially interesting events for a human analyst to follow up on. The disconnect it took for someone to blindly trust the program as a target acquisition system boggles my mind. Someone doing donuts in a parking lot should not be automatically targeted with hellfire missiles, yet in practice that is how they were using the tool. I still lose sleep over it.

I'm sorry for your pain.

This is a good lesson for young people reading this. If you work for the US military, you are not working for Defence.

Your work will be used against innocent civilians in far away lands.

Not sure about the downvotes, this is correct even if not pleasant to read (people need to be heroes of their little bullshit life story and all that).

It may be an afterthought for US citizens living in US, but its a deadly weapon pointing to 95% of the mankind and people wielding it are often shit to be polite. We the rest do care.

I mean, this site is full of people that develop the weapons used to commit atrocities and genocides (both physical and digital ones) so I am not surprised about the downvotes..

I can't wait for such a system being deployed stateside.

If that's what it takes for Americans to start caring about the harm their military is inflicting on the world, sure.

I believe its called the second amendment.

I think you're conflating the impact of a small bubble on a narrow market space with our current situation. The current bubble is larger than anything ever seen before, and is now encompassing nearly the entire economy.

New datacenter builds are so far along the curve of diminishing returns it's absurd. No one is going to want to pay to run a datacenter that costs 10x as much to run for the same compute.

And the problem is with all these "freed" resources, the entire pipeline will be affected. No one will want to buy any new silicon if they can buy a B200 for $1000. We could potentially see a decade or more of stagnation in the chip sector, or even significant regressions in capabilities as foundries are shut down due to lack of demand.

The impact of what is coming scares me to my core. I don't think we're going to bounce back from this any time soon.

I think actually it's quite the opposite and you're the one conflating the impacts on a narrow market space with the larger economy. Right now, AI dominates growth in the stock market and demand for chips, but it is by no means encompassing the whole economy. Neither the stock market nor the chip sector are the entire economy. My grocery store isn't going to go out of business if Anthropic does.

A lot of investors may lose money as a result of the bubble bursting, but that does not mean the underlying asset will be forever worthless, just that it didn't provide sufficient returns sufficiently quickly to justify the upfront investment for the investors that funded it, at the time they made that investment. A different investor who could afford to ride out a period of reduced demand might have an entirely different experience.

Imagine you take out a five year loan to buy a truck to deliver packages, but then for the first two years you operate it, gas prices are elevated so you struggle to make the payments on your loan and end up not making as much money as you had hoped to originally, perhaps even to the point you need to declare bankruptcy and sell of the vehicle. But if not and then gas prices drop back down and demand shoots up for the last three years of the loan, you could then make up the difference. If the first two years drive you into bankruptcy, that is difficult for you, but amortized over the entire five year loan period, the truck may actually have been a profitable investment for someone who could have afforded to ride out the first two years. And just because you go bankrupt doesn't mean the truck stops being a valuable asset, it's just that you don't end up benefiting personally from that value because your timing was bad. From a macro perspective, the overall economy doesn't suffer, except to the extent that it might have been more efficient to invest the capital that went into procuring the truck elsewhere during those first two years. But only possibly and only on the margins, because the truck remains a profitable investment over the course of its entire lifetime.

Don't confuse the success or failure of individual investors or businesses with the success or failure of the overall economy. Current data center build outs premised on fanciful projections of demand for LLMs may end up not being profitable in the short term while still being profitable over the entire productive lifetime of the asset, if sufficient demand is found elsewhere or if demand for LLMs picks up later. Similarly, I anticipate at worst we will see chip prices plateau for a while if there is a pullback in LLM demand, but we won't see them fall and they will continue to rise over the longer term as more demand is generated elsewhere.

As one small example: we have barely begun to scratch the surface of what we can achieve with robotics. Think about the demand for video processing if you have tens of thousands of robots stocking shelves in supermarkets generating video all day long. On board processing will of course be the obvious primary demand for chips, which doesn't benefit data centers, but central processing of video to extract useful data from the entire fleet will generate demand for data centers. As will large scale training jobs. Now multiply that thinking across the entire scope of industries where robotics may be useful for replacing human labor, and you're talking about an extremely significant amount of valuable computational work.

Yeah but general purpose data centers that allow hobbyists to spin up an app or a service easily is exactly that: general purpose. AI data centers are built for the very specific kind of math that LLM's require. Even the GPU's can't even be used for something like cloud gaming, which isn't very popular anyways.

That math is very much general purpose. It's just linear algebra under the hood, and linear algebra is a wildly useful toolkit with an unbounded set of potential applications, including many applications other than LLMs. That's how something developed for gaming evolved into something used for text processing and generation. For example, we have barely begun to scratch the surface of what we can achieve with robotics, and I have no doubt that further advances in robotics will generate an enormous amount of computer vision work that GPUs are extremely well designed to handle.

"a niche gaming product would enable an industrial revolution level technological leap forward."

something has to true to be surprising. your statement is false and nonsese

Bullshit. Even if you don't think LLMs alone are enabling that level of advancement, GPUs in general very much are. They are the critical enabling technology underlying self-driving cars, autonomous drones, camera-based robotics controllers, faster/better drug discovery, faster/better modeling protein structure and design, ML weather forecasting, materials discovery, and many other applications.

The eventual utility of the Internet did not stop a lot of people losing a lot of money in the dotcom crash.

Fair, but my point is that a data center is not pets.com. They are a much more flexible asset, and are even more flexible than AI itself. Pets.com failed because the physical delivery infrastructure and consumer buying habits that support Chewy today did not yet exist and took years to create. Data centers can be repurposed from supporting the current generation of LLM-based AI models to an infinite variety of use cases. Pets.com could only ever hope to deliver pet food.

And Amazon.com could only ever hope to deliver physical books?

Anyway, it's clear an AI data center has utility for crunching AI inference, and that there is and will continue to be demand for AI inference. The trillion dollar question is whether you can make money doing this, and so far the answer is no.

https://isaiprofitable.com/

You are not disproving but rather reinforcing my point. Amazon survived the dotcom bubble precisely because they could hope to do more than deliver physical books. It was a more flexible, generally applicable business model than pets.com, and they started out targeting a market more susceptible to ecommerce conversion than pet food delivery proved to be. Similarly, a company selling a product that can only be used for the limited use cases we have found for chatbot-style text generating LLMs will be more likely to fail than a company selling something that is more generally useful. Data centers are more generally useful than LLMs, and the companies building them are more likely to survive than those whose fortunes are pinned entirely to unrealistically high expectations for replacing white collar workers with LLMs.

Data centers aren't just physical buildings though - the vast bulk of data centre costs is the racks of servers in them, and those are being bought and installed on credit at very high costs.

Its not about "is a computer thoeretically useful" its about if it's going to deliver value to cover the cost of installing it in the first place over it's expected lifetime.

You aren't substituting generic compute: you're substituting highly optimized GPU dense server space. The application is more limited then it looks.

Do you mean is there anyone who thinks that the future is NOT local models anyway? Since that seems to be the gist of your other points

Yes that’s what I meant thanks for clarifying. I think local models are the future but I don’t see anyone serious arguing it. If it were true these city sized data centers will be graveyards in 5-10years

Humanity might be able to, humans cannot

If we (as a species) can genetically modify embryos to make our offspring more athletic, more intelligent, more <whatever>, is that human-RSI?

That's not a gotcha question.

I'm just genuinely curious if that would meet the definition of RSI.

I'm shocked Netflix hasn't just settled to keep this quiet, because the entire situation makes them look not only like an abusive, discriminatory employer but even worse, horrifically incompetent at being abusive and discriminatory. Plenty of companies treat their employees like shit, that's not news. But any competently managed company knows how to do so without incurring massive legal liability like this, and it speaks to a remarkable lack of professionalism and basic competence at the leadership level and within the legal and HR teams.

If the accusations are true as reported, their in-house counsel fucked up terribly by assuming all ketamine use is illegal and recreational without bothering to conduct the very cursory investigation required to determine that it is in fact a perfectly legitimate option for treating treatment-resistant depression. That single error drove them to fire an otherwise exemplary employee for an obviously illegal reason in a manner that screwed him out of significant compensation including severance.

This was an obvious, unforced error by their legal team. Competent lawyers and competent corporate leadership would recognize that the relatively small amount of money they may save by litigating this issue is not worth the damage it is very predictably inflicting on their corporate culture and public brand. The lawyer who publicly admitted that ketamine factored into their decision should be immediately fired because he is demonstrably a terrible fucking lawyer, and he seems to be representative of the quality of legal advice Netflix is receiving, because any attorney with half a brain would have immediately recognized both the underlying legal error that incurred liability for the company and the business context that makes immediate heavily NDA'd settlement the optimal strategy for defending, even were the original legal error not so egregious.

My main takeaway from this situation is that Netflix is managed by remarkably incompetent leaders who are advised by even more remarkably incompetent lawyers.

You can't have an HR policy that discriminates against people for receiving mental health treatment from a physician, that is plainly illegal under multiple statutes both federally and in the state of California. This was not a situation where he was on ketamine while at work, and there are no accusations that his ketamine use even affected his work at all, even incidentally or indirectly. Even if it had, Netflix would be required to make a reasonable accommodation for him. But we know it did not affect his work because he received treatment in 2022 and Netflix only found about it in 2026 after he disclosed the treatment. Besides, he is a movie executive, it's not like he is operating heavy machinery. Ketamine is unlikely to have any effect on him that would in any way expose Netflix to liability.

Netflix seems to have simply assumed that all ketamine use is illegal and recreational, without bothering to conduct even the most cursory of investigations, which would have revealed that ketamine is in fact a clinically acceptable treatment for treatment-resistant depression. This is what happens when you have utterly incompetent idiots for in-house counsel and HR.

the case of unstoppable force (HR) meeting an immovable target (FMLA, ADA, EEOC, FEHA)

Yes, it seems likely they violated California's Fair Employment and Housing Act (FEHA) and the federal Americans with Disabilities Act (ADA) based on the details I have seen so far. The EEOC’s mental-health guidance states both that an employee may choose to discuss his condition with coworkers and that the employer may not discriminate against him for doing so. https://www.eeoc.gov/laws/guidance/depression-ptsd-other-men...

That's only guidance, not binding law in and of itself, but it means that the employer can only discourage employees from discussing their mental health if they do so in a neutral manner that applies equally to all employees in a non-discriminatory manner, which was self-evidently not the case here.

Notably, the EEOC guidance standards are not wholly unlimited. Employers can take action against an employee if their discussion of their mental health is repeatedly inappropriate or disruptive. e.g., if someone is constantly cornering coworkers to trauma dump suicidal thoughts on them or something like that. But only in truly extreme cases where an accommodation is impossible (e.g., if they are violent) or if they fail to adjust their behavior after receiving feedback that it has become unacceptable. A single incident that does not appear to have even made any other employee uncomfortable certainly doesn't suffice.

Netflix's only real defense here would be to argue that they would have fired him anyway without the disclosure, in response to other issues. That is a challenging defense to make given the admission their lawyer made that the ketamine treatment factored into their decision to fire, but the ADA and FEHA require "but for" causation, i.e. he is protected from firing if they would not have fired him but for the admission of ketamine treatment. I doubt they will succeed given the relative triviality of their other accusations (somewhat excessive profanity in a context where a fair amount of profanity was considered acceptable) and the entirely inoffensive party trick (at least in a context where the CEO has been repeatedly photographed drinking alcohol at company events and the alcohol at this event was provided by the company). I seriously doubt they will be able to point to any similarly situated employees that Netflix has previously fired solely for profanity or consuming alcohol that the company provided to them (and from a PR perspective that would almost be worse for them to admit). Frankly, the profanity feedback also seems like the very common scenario where a manager is required to provide regular feedback but can't think of anything constructive to say because the employee is a high performer, so they reach for something funny and minor just to check the box. In the absence of extensive complaints from coworkers, it is unlikely to overcome the company lawyer's outright admission that the ketamine use was a factor in their decision.

What appears to have happened here is Netflix has awful in-house counsel and/or HR who utterly failed to carry out their duties with even minimal competence. They appear to have simply assumed that all ketamine use is automatically illegal recreational drug abuse without bothering to investigate whether that is true in general or in this particular case before escalating to the most extreme possible reaction. Those employees are the ones who should be fired, not only because of the gross incompetence it takes to so egregiously violate America's otherwise absurdly weak legal protections for workers' rights, but because they did so in a manner that seems practically designed to permanently destroy employee trust in the company while inviting unwelcome public scrutiny of their potentially discriminatory employment practices.

This is a fairly straight forward discriminatory-causation story: Netflix invited vulnerability and thus potential mental health related disclosures, learned of a psychiatric history, reframed treatment as drug misconduct, and then expressly treated it as a termination factor. It's honestly pretty rare to see such an obvious example of this kind of discrimination, usually companies do a better job covering it up with a pretext, and usually their *lawyers* aren't so unbelievably fucking stupid as to admit publicly to the discriminatory decision.

That is a hopelessly naive description of lobbying, to the point that I assume you are being intentionally disingenuous. Lobbyists commonly funnel money to politicians via fundraiser bundling, donations to outside spenders like PACs, and other mechanisms that get around the weak remaining restrictions on direct campaign contributions that SCOTUS has not yet gutted. A lobbyist may not donate directly to a politician's campaign, but they get a group of their friends together for a fundraiser to effectively donate on their behalf (bundling). Or they donate to outside groups who then spend to support the politician's campaign, e.g. by funding advertising that benefits the candidate or hurts their opponent. Like when AIPAC sinks millions into ads against progressive candidates, that's tightly connected to their lobbying on behalf of Israel and you better believe that the politicians benefiting from that spending are well aware of how it is connected to their willingness to adopt the lobbyists' positions.

I stopped cooking with nonstick pans over 15 years ago, and the only thing I've found that can consistently be a bit tricky is eggs, but really only if the pan isn't hot enough yet when you add the butter or oil, you don't use enough butter/oil, you try to scramble the eggs in the pan, or you try to lift the eggs too soon. So arguably it takes a little more attention to detail to cook eggs on stainless steel, but in general I really don't think there is anything that requires nonstick. But even for the limited list of items that are easier to prepare on nonstick, it's usually just a matter of learning how to adjust your technique to match the cookware.

I made the switch to stainless steel and cast iron pretty much on a whim after I first read about Robert Bilott's lawsuits against DuPont, at which point I discovered it made almost no difference to my quality of life at all to avoid nonstick. To such an extent that I have never been even slightly tempted to switch back, despite honestly not feeling a particularly strong motivation to avoid nonstick in the first place. I've just always figured cooking with steel is an entirely painless way to reduce my potential exposure to dangerous chemicals, so even if it ends up being pointless, I haven't lost anything of value anyway.

Eggs stick if the pan temperature isn't right or if there's not enough lube. User error or skill issue. Here's how to do it: buy good pans, turn on pan to ~medium, add oil and rub in with cloth, add more oil (or butter, butter is better with eggs), when pan is hot add eggs. This works with good quality (all clad) or heavy (cast iron) pans.

Personal opinion - deep fried eggs are goated. It's either runny scrambled (pretty hard on cast iron/stainless) or deep fried. Anything in-between is incompetence.

Runny scrambled (French omelet) isn't too bad on carbon steel IME. And certainly possible on cast iron, since that was the type of pan used when Carême wrote his cookbook. By Escoffier's time carbon steel was more common. Neither of them is famous for having invented a time machine, both predate non-stick cookware, and they're the most famous French chefs in history.

French omelet is completely different thing. I do use my cast iron for this, it's okay if you go hard with silicone spatula. But still, more work than nonstick. I keep meaning to get carbon steel!

By "runny scrambled" I mean a folded omelette with a runny center, just-set exterior, and no browning. Classic French style, not just slop on a plate.

> Add oil and rub in with cloth

Do what now?

I don't get why decoupling insurance from employment isn't a bigger part of the healthcare reform discourse. It's the most obvious, easiest, lowest hanging fruit for improving the US healthcare system, and almost uniquely politically acceptable to both Democratic and Republican voters. Decoupling also doesn't entail disrupting any of the established incumbents (pharma, AMA, insurance companies, etc.) who typically kill any reform. The policy is also arguably market-oriented, or at least is not a form of government intervention that would otherwise offend Republicans, yet benefits individuals over large companies so can appeal to Dems too.

The Republicans have utterly failed to offer *any* alternative at all to Obamacare after 16 years, which is insane for many reasons, but especially considering that they could have put decoupling insurance from employment at the heart of an alternative (even more than the ACA) market-oriented reform. Letting individuals choose their own insurance would put more pressure on insurance companies to actually deliver quality care because, unlike employers, individuals are highly sensitive to the quality of care they receive and are more likely to switch insurers if the insurer unreasonably denies claims or otherwise fails to deliver a decent product. On the other side, individuals are also more sensitive to costs so with greater control over their plan choices would be more able to pick plans that are structured to match their anticipated costs.

In other words, individual insurance would improve the functioning of the market, which should in theory appeal to Republicans. Instead, they only focus on making individuals pay more at the point of care, on the theory that cost-sharing discourages insured people from over-consuming healthcare. Yet they don't think about how the current system makes it impossible for individuals to be truly cost sensitive when selecting plans.

I would guess the Republicans haven't adopted the idea because they are not really in favor of free markets, they are in bed with big business, and employer-provided insurance tends to favor larger companies that can demand better terms from insurers, that can amortize the HR overhead over a larger number of employees, and that tend to benefit disproportionately from lower labor mobility compared to small business (people are more likely to be involuntarily locked into a big company job for security than at a small business, and it's harder for people who are worried about maintaining quality healthcare access to move to a startup, reducing the competition experienced by the bigger companies).

I’m likewise baffled. Healthcare spending is a major cost for all employers. Decoupling healthcare from employment would theoretically remove this cost immediately. Shouldn’t pro-business groups have long been lobbying for this?

The fact that they are not tells me there must be some other financial incentives at work that are unclear to me.

I think a lot of people want to decouple insurance from employment but the mechanics would be tricky.

Ideally all business would take what they pay for insurance and stick it in the employee's paycheck and then the employee would buy insurance with that money. Reality says, a lot less people end up having insurance if you do it that way. Maybe force companies who once offered insurance to fund an employee HSAs and change the HSA rules so it can be used to buy insurance. It would take some thoughtful design to actually pull off the transition while keeping people on insurance.