Money is not given to good ideas (though, it doesn’t hurt). Money is given to friends. If you look at how VC (or really any network) funding circulates, it’s just people who are allowed to enter that circle and money just flows between them constantly. On one hand, you have trusted people who you are willing to give money, on the other hand, this inherently creates a clique.
It reminds me how the Bohemian Club’s slogan, “Weaving Spiders Come Not Here” is a bit farcical given that it is impossible for the club members not to engage in commerce.
I really don’t understand the structure that Mozilla has setup here where Thunderbird lives under their for-profit branch, but is so dependent on donations. I remember reading they were considering charging for Thunderbird services when the move was announced 6 years ago, but as far as I can tell, nothing has happened and they’re still desperate for funding.
Now not only does it still need donations, the tax exemption for donors has evaporated. Great.
Nix provides declarative, reproducible builds. So, ostensibly, if you had your build system using Nix, then some of the issues here go away.
Unfortunately, Nix is also not how most people function. You have to do things the Nix way, period. The value in part comes from this strong opinion, but it also makes it inherently niche. Most people do not want to learn an entire new language/paradigm just so they can get this feature. And so it becomes a chicken and egg problem. IMHO, I think it also suffers from a little bit of snobbery and poor naming (Nix vs. NixOS vs. Nixpkgs) which makes it that much harder to get traction.
Based on what? A lot of this is vibes and FOMO; just like any economic bubble.
There is no objective evidence of anything you’ve said. It isn’t even clear if AI has contributed positively to global economic growth. It reminds me a lot of the late 90s and the dot-com mania. Slapping a domain on a commercial would make your stock go up even if there was no substance to any of it.
The real shame is this mania drowns out serious, practical use cases because when the bubble collapses, the market will throw the baby out with the bathwater.
Regardless they are getting that revenue through genuine demand for their product. It’s not like they are selling back some commodity product, billions are being spent on model outputs.
I think anyone who has used Opus 4.6 can see what is causing this demand. It is genuinely “smart” in the sense that it can work its way around non-trivial coding problems.
I don't see why tokens/$ would suddenly stop dropping. Maybe this is the first time the cost of compute will plateau, but do have any reason to think so?
There is a strong suspicion, especially of people who are skeptical of AI, that the actual price is being severely subsidized. The sense is that it’s an extreme version of growth before revenue. It is questionable if the true cost of training and inference make any of this worthwhile once Anthropic/OpenAI need to stand on their own and make money.
Imagine you open a cookie shop and you are VC funded, so you charge 5¢ for a cookie to attract people.
- Your real cost is $20/cookie. $15 for the fancy retail packaging and presentation, $5 for baking each cookie.
- You get lots of attention, strong profits and go public.
- VC funding is gone so, now instead of charging 5¢, you now need to charge $25 in order to not be in the red.
One of the reasons people think this is the shenanigans that Anthropic is currently playing, quietly tweaking the behavior of Claude Code and whatnot without really telling people. You can see lots of comments online about Claude Code randomly feeling dumber before Anthropic engineers admit they are messing with it.
Imagine you are on the $200/month Max plan. If the sustainable cost of this is several orders of magnitude higher, would enough current users pay something like $3,000/month for what we currently have?
Sure, yeah, I saw grubhub happen too... but this is compute, not cookies. It gets cheaper.
I don't even get what "skeptical of AI" means. We made AI, many companies reliably teach computers every spoken language. I perform my white collar job with a massive AI multiplier to my productivity.
I'm typing this on a machine comparable to Japan's Earth Simulator, a $350M supercomputer.
There are no talks or anything. Iran has no incentive to negotiate with a party as unreliable as the US is under Trump. I would literally negotiate with a dead opossum before I would continue to negotiate with Witkoff and Kushner.
I mean, as much as I don’t like the Iranian government, put yourselves in their position. You have the US and Israel literally leveling the equivalent of Balfour or the White House and taking out other government officials in a decapitation strike that failed, but killed off all of the moderates. The government is then replaced by hardliners who see this attack as existential. You have little to lose at this point, so you go for broke.
Since the US seems unwilling to put boots on the ground, cannot form a coherent reason for any of this and is lead by a man who is unable to accept that he can commit errors, it degrades into a war of attrition and, in the case of Trump, influence peddling since it is clear that Israel and the Saudis would like to see Iran wiped off the map and all Trump cares about is how he can internalize it as yet another reason why he is a victim and entitled to the Nobel Peace Prize.
IMHO, I think there is tremendous pressure to, at the very least restore the Strait of Hormuz as an international waterway not subject to Iranian control or tolling, but that’s an after-the-fact thing. I think Trump simply thought it would be an easy win and play well on TV. I suspect what will happen is the US pays a massive indemnity/bribe to Iran, Iran agrees to not contest control of the Strait of Hormuz and the US looks like morons which Trump will internalize as a win that nobody will believe except himself.
> There are no talks or anything. Iran has no incentive to negotiate with a party as unreliable as the US is under Trump. I would literally negotiate with a dead opossum before I would continue to negotiate with Witkoff and Kushner.
> Iran, while rejecting all the plans presented by the enemy, formulated a 10-point plan and presented it to the US side through Pakistan, emphasizing the fundamental points such as controlled passage through the Strait of Hormuz in coordination with the Iranian armed forces, which would grant Iran a unique economic and geopolitical position, the necessity of ending the war against all elements of the axis of resistance, which would mean the historic defeat of the aggression of the child-killing Israeli regime, the withdrawal of US combat forces from all bases and deployment points in the region, the establishment of a safe transit protocol in the Strait of Hormuz in a way that guarantees Iran's dominance according to the agreed protocol, full payment for the damages inflicted of Iran according to estimates, the lifting of all primary and secondary sanctions and resolutions of the Board of Governors and the Security Council, the release of all of Iran's frozen assets abroad, and finally the ratification of all of these matters in a binding Security Council resolution. It should be noted that the ratification of this resolution would turn all of these agreements into binding international law and would create an important diplomatic victory for the Iranian nation.
> Now, the Honorable Prime Minister of Pakistan has informed Iran that the American side, despite all the apparent threats, has accepted these principles as the basis for negotiations and has surrendered to the will of the Iranian people.
> Accordingly, it was decided at the highest level that Iran will hold talks with the American side in Islamabad for two weeks and solely on the basis of these principles. It is emphasized that this does not mean an end to the war and Iran will accept an end to the war only when, in view of Iran's acceptance of the principles envisaged in the 10-point plan, its details are also finalized in the negotiations.
> These negotiations will begin in Islamabad on Friday, April 11, with complete distrust about the US side, and Iran will allocate two weeks for these negotiations. This period can be extended by agreement of the parties.
When you use words like "decapitation strike that failed, but killed off all of the moderates," what do those words mean to you? With all due respect, I don't really get the Internet brain way of thinking of things. What decapitation failed? I guess, if you mean, there are still Islamic Revolution people in charge, I still can't see the point. When you say "failed" that would imply that they were literally attempting to kill literally every single member of the government at once. I don't think anyone serious would think that. Also, "failed?" I can't recall ever a decapitation happening so swiftly or so massively within the first few hours of a conflict. Also, the meat of what I wanted respond to was this idea of "killing the moderates." I get that most people here think the West and America is evil or whatever but the idea the Ayatollah and top members of the IRGC were moderate is just an affront to morality. The same people think that Trump is Hitler for doing things that 90s Democrats agreed with (even ones currently serving), would hold vigils for a truly monstrous regime. This is like some Billie Eilish "no one is illegal on stolen land" type stuff. We are talking about brutal executions for no reason at all.
decapitation was intended to result in regime change, but instead showed that the iranian system is perfectly capable of peaceable changes in power. what particularly failed is that the people the US wanted to champion as the new leaders of iran were also killed in the decapitation.
you can compare against the successful decapitation from christmas, where the US removed maduro, and championed rodriguez and now takes a cut of all venesuelan oil sales.
i think there's a reasonable argument that the ayatollah was a moderate, in a much more militant government. He's the guy that was making sure iran never built a nuke, and by observation, iran stood down after each attack the US/israel did on iran up until he was gone
"no one is illegal on stolen land" is perfectly reasonable - the american government has no actual legitimacy to control who comes and goes from land that doesnt belong to it. the various tribes do. its impractical in that the US genocided the legitimate owners and took it over by force, but its still the right and just end view. the US gets to kick people out of certain borders because it did a ton of brutal executions
> I get that most people here think the West and America is evil or whatever but the idea the Ayatollah and top members of the IRGC were moderate is just an affront to morality.
I really don’t understand this logic. I find it rather myopic and based on one’s own pain. Everything is relative, unfortunately. The idea that I would in any way condone or argue that the Iranian regime is not culpable of its own massive war crimes, grifting and other crimes against its own people is…bizarre. I am well aware of the crimes of the Iranian regime and look forward to the day it is removed, but I don’t think this is it. Even Trump admits that they killed off all of the people they thought would be more amenable to work with the US which is just a level of incompetence I can’t fathom, but here we are.
Unfortunately, in practice, moral absolutism does not exist in international relations. The evidence is right in front of your face of this fact. We could go through the litany of crimes against people that we (the US) have condoned or facilitate or been unresponsive to. The folks in Beijing have also committed unspeakable acts against their own people and others, so why aren’t we bombing them right now? Why Iran right now? Haiti is a failed state nobody seems interested in caring about. We failed to stop a genocidal massacre in Rwanda…
> When you say “failed” that would imply that they were literally attempting to kill literally every single member of the government at once.
I literally believe that Trump thought this given that he openly admitted he ignored the military and intelligence agencies telling him that this was a terrible idea. I agree that nobody rational would think this, but I argue that Trump never lies even when he says he is joking. He literally thinks as POTUS he can do whatever he wants.
Anthropic needs to show that its models continually get better. If the model showed minimal to no improvement, it would cause significant damage to their valuation. We have no way of validating any of this, there are no independent researchers that can back any of the assertions made by Anthropic.
I don’t doubt they have found interesting security holes, the question is how they actually found them.
This System Card is just a sales whitepaper and just confirms what that “leak” from a week or so ago implied.
I think it’s also important to add the context that Broadcom’s CEO, Hock Tan, went on CNBC in October and had a vacuous conversation with Jim Cramer about their OpenAI “deal” at the time [0]. Nothing of substance was said, it was just endless loops about the opportunity of AI. It is now 6 months later and there has been nary a peep from Broadcom about any updates.
I think Anthropic is a more grounded company than OpenAI because Sam Altman is insane, but it is still playing the same game.
Because all the variables that go into performance / efficiency measurement of a model (processing power, algorithm efficiency, parallelization, etc) boil down to cost per token input and token output. And the tangible cost for a datacenter is power consumed. Of course, amortized capex costs are also part of the game.
It's easy to think about. Google reported a global average power consumption of 3.7GW in 2024, so you can think of this deal as representing an expansion of something like 10-15% of that 2024 baseline, if you assume 50% capacity utilization.
There's at least a decent argument to be made that the limiting factor is actually the physical silicon itself (at least at cutting-edge nodes) not really the power. This actually gives AI labs an incentive to run those specific chips somewhat cooler, because high device temperatures and high input voltages (which you need to push frequencies higher) might severely impact a modern chip's reliability over time.
Power is the limiting layer above physical chips. You can add more chips and them at lower clock or add more efficient chips later on, but you can't really change the power of a data center easily.
It will nonetheless be vastly cheaper to build a new datacenter and arrange for powering it than to fab the amount of leading-edge chips and compute systems that are going to ultimately eat that power. So the chips themselves are still the meaningful constraint.
Surely, there should be some more critical questions posed by why just buying a bunch of GPUs is a good idea? It just feels like a cheap way to show that growth is happening. It feels a bit much like FOMO. It feels like nobody with the capital is questioning whether this is actually a good idea or even a desirable way to improve AI models or even if that is money well spent. 1 GW is a lot of power. My understanding is that it is the equivalent to the instantaneous demand of a city like Seattle. This is absurd.
It feels like there is some awareness that asking for gigawatts if not terrawatts of compute probably needs more justification than has been proffered and the big banks are already trying to CYA themselves by publishing reports saying AI has not contributed meaningfully to the economy like Goldman Sachs recently did.
kinda complicated though when you consider it fully. Power consumption only measures the environmental impact really, we come up with more clever ways to use the same amount of power daily.
it's kind of like an electrical motor that exists before the strong understanding of lorentz/ohm's law. We don't really know how inefficient the thing is because we don't really know where the ceiling is aside from some loosey theoretical computational efficiency concepts that don't strongly apply to practical LLMs.
to be clear, I don't disagree that it's the limiting factor, just that 'limits' is nuanced here between effort/ability and raw power use.
"Do you realize that the human brain has been liken to an electronic brain? Someone said and I don't know whether he is right or not, but he said, if the human brain were put together on the basis of an IBM electronic brain, it would take 7 buildings the size of the Empire State Building to house it, it would take all the water of the Niagara River to cool it, and all of the power generated by the Niagara River to operate it." (Sermon by Paris Reidhead, circa 1950s.[1])
We're there on size and power.
Is there some more efficient way to do this?
Maybe it's just because the specifics on FLOPs are more complicated, especially given how many different floating point formats are floating around in ML. Even NVIDIA has like 6 different FLOPs numbers on their GPUs nowadays.
And you know Nvidia can't be constent with one format for FLOPs within a single graph, 1,000,000x faster but comparing FP32 to FP8 or NVFP4 and acting like it's the same.
Some of it might be market-signaling to the broader energy industry: "hey would you PLEASE build more power plants and power lines? Look at all this money we have, we will pay for it!"
I thought the sales pitch of all of this is that the AI was supposed to relieve people from having to do a bunch of annoying bootstrap coding and to do it in a way that we could extended easily.
I have a subscription to Claude Code and despite my skepticism, it has been pretty good at just getting a goofy PoC thing going. When I look at the code, it’s usually insane unless the prompt was so narrow and specific like about writing a function that does one thing and only one thing.
Outside of small, personal projects, I am still really uncomfortable at having agents run wild. I see the result, and then I spend a bunch of time having to gain the context of what is going on, especially if I ask it to implement features in spaces I have general knowledge, but not expertise. So, the problem remains the same. These things still need handholding by people who understand the domain, but having people become glorified PR reviewers is not an acceptable path forward.
Arguing that there is lots of bad production code kinda avoids the actual issue that is going on here. Yes, a lot of sloppy code can and has been written by people. I’ve seen it myself, but it feels like the actual thing is that, we are now enabling that at scale and calling it “abundance” when instead we are really generating an abundance of completely avoidable security holes and logic errors.
No, it is isn’t and saying it is here by using vague goalposts does not make AGI show up. I will agree we have been unable to define what it is, but we can’t even figure out why humans have consciousness right now.
1) Still unreliable at logic and general inference: try and try again seems to be SoTA...
2) Comically bad at pro-activity and taking the right initiative: eg. "You're right to be upset."
3) Most likely already reaching the end of the line in terms of available good training data: looking at the posted article here, I would tend to agree...
The problem is that LeCun was obviously wrong on LLMs before. You have to take what he says with the caveat that he probably talks about these in a purist (academic) way. Most of the "downsides" and "failures" are not really happening in the real world, or if they happen, they're eventually fixed / improved.
~2 years ago he made 3 statements that he considered failures at the time, and he was quite adamant that they were real problems:
1. LLMs can't do math
2. LLMs can't plan
3. (autoregressive) LLMs can't maintain a long session because errors compound as you generate more tokens.
ALL of these were obviously overcome by the industry, and today we have experts in their field using them for heavy, hard math (Tao, Knuth, etc), anyone who's used a coding agent can tell you that they can indeed plan and follow that plan, edit the plan and generally complete the plan, and the long session stuff is again obvious (agentic systems often remain useful at >100k ctx length).
So yeah, I really hope one of Yann, Ilya or Fei-Fei can come up with something better than transformers, but take anything they say with a grain of salt until they do. They often speak on more abstract, academic downsides, not necessarily what we see in practice. And don't dismiss the amout of money and brainpower going into making LLMs useful, even if from an academic pov it seems like we're bashing a square peg into a round hole. If it fits, it fits...
> we can’t even figure out why humans have consciousness right now.
My uneducated guess is that it just means we save/remember (in a lossy way) inputs from our senses and then constantly decide what to do right now based on current and historical inputs, as well as contemplated future events.
I think the rest of our body greatly influences all of that as well, for example: we know running is healthy and we should do it, but we also decide not to run if we are busy, feel tired, or are in pain etc.
"the hard problem of consciousness" is not about "what are we conscious of", but rather: how is it possible to be conscious (i.e. experience qualia) at all?
I think we agree - we have arbritrary goalposts regarding AGI and they have been met. WE don't know what we consider to be "the big changing moment" and that moment is hard to define because we don't have a good definition of it when we talk about ourselves even.
So the convo becomes - what is that "thing" and do we need to draw similarities between "it" and our own intelligence.
Why would it specifically be job-displacement-via-LLM's that signal AGI? Why not job-displacement-via-automated-robot? or job-displacement-via-office-technology?
That seems like a pretty dangerous place to set your goalposts. Don't you think we need to see what's coming and figure out how to deal with it before widespread job destruction starts?
It's what everyone other than pedantic jerks think AGI means. Read a book. Watch a movie. Every depiction of AGI is essentially a depiction of a human. Sometimes a human without substantial emotions. That's sentience.
The focus of AGI is on achieving human equivalence in cognitive tasks, or to surpass it ("intelligence"). That's where the money and the research is. Making a stupid machine that happens to be aware ("sentient") isnt the goal.
It reminds me how the Bohemian Club’s slogan, “Weaving Spiders Come Not Here” is a bit farcical given that it is impossible for the club members not to engage in commerce.
reply