By your analogy, the developers of stockfish are better chess players than any grandmaster.
Tool use can be a sign of intelligence, but "being able to use a tool to solve a problem" is not the same as "being intelligent enough to solve a specific class of problems".
Im not talking about this being the "best maze solver" and "better at solving mazes than humans". Im saying the model is "intelligent enough" to solve a maze.
And what Im really saying is that we need to stop moving the goal post on what "intelligence" is for these models, and start moving the goal post on what "intelligence" actually _is_. The models are giving us an existential crisis on not only what it might mean to _be_ intelligent, but also how it might actually work in our own brains. Im not saying the current models are skynet, but Im saying I think theres going to be a lot learned by reverse engineering the current generation of models to really dig into how they are encoding things internally.
If making money on inference alone was possible, there would be a dozen different smaller providers who'd be taking the open weights models and offering that as service. But it seems that every provider is anchored at $20/month, so you can bet that none of them can go any lower.
> If making money on inference alone was possible, there would be a dozen different smaller providers who'd be taking the open weights models and offering that as service.
There are! Look through the provider list for some open model on https://openrouter.ai . For instance, DeepSeek 3.1 has a dozen providers. It would not make any sense to offer those below cost because you have neither moat nor branding.
Maybe, but arguably a major reason you can't make money on inference right now is that the useful life of models is too short, so you can't amortize the development costs across much time because there is so much investment in the field that everyone is developing new models (shortening useful life in a competitive market) and everyone is simultaneously driving up the costs of inputs needed for developing models (increasing the costs that have to be amortized over the short useful life). Perversely, the AI bubble popping and resolving those issues may make profitability much easier for the survivors that have strong revenue streams.
You need a certain level of batch parallelism to make inference efficient, but you also need enough capacity to handle request floods. Being a small provider is not easy.
Why would the open weights providers need their own tools for agentic workflows when you can just plug their OpenAI-compatible API URL into existing tools?
> when you can just plug their OpenAI-compatible API URL into existing tools?
Only the self-hosting diehards will bother with that. Those that want to compete with Claude Code, Gemini CLI, Codex et caterva will have to provide the whole package and do it a price point that is competitive even with low volumes - which is hard to do because the big LLM providers are all subsidizing their offerings.
You are vastly overestimating people's willingness to deal with bullshit, when the product does not have a real lock in.
It would be incredibly easy to have a company offering their ChatGPT over WhatsApp or iMessage, and get people to start using it instead of an ad-ridden GPT app.
The funny thing is that make a big deal about blocking Brave on "ethical" grounds, but don't e tend the logic to Chrome, Edge or Safari. Talk about punching down/virtue signaling.
They outline a very specific behaviour that Brave engaged in but Chrome, Edge, and Safari do not. Brave was engaging in fraudulent behaviour, wherein it posed a fake donations scheme to users of the browser under the guise of supporting website owners with their implicit but nonexistent consent, and in actuality took the money for itself. Brave then also specifically and publicly singled out Lobsters in an issue. Lobsters devs do not want to spend dev time engaging with scammers operating in bad faith. Seems fair to me.
Allegedly. This, the "Brave was putting ads of their own on other pages" and "adding the referrer code for Binance" stories get thrown around like they were (a) are all huge sources of profit (b) carried on with malicious intent and (c) on par with the BILLIONS of dollars in ad fraud that goes around and Google so conveniently turns a blind eye.
I don't particularly care how much money Brave made off the scheme. If Brave put my name and picture on an advertisement shown to Brave users, said that I was soliciting donations and would receive the money, and then took the money, that is immediately far more personally offensive than virtually anything Google does. I was not myself actually affected by this, but it's incredibly easy for me to understand why someone would want nothing to do with Brave.
Also, it's basically a given that Brave is not in a position to generate billions through such a scheme. It simply doesn't have the market share for that. If Chrome did the same thing that Brave did, they probably would generate billions. It is equally unethical either way.
Moreover, Google has an effective monopoly. Even if you wanted to protest Chrome, you can't do so without effectively shutting down your website. Chrome coerces consent into whatever they do. Brave does not have that power. You describe that as punching down, but just because Chrome has the capability to coerce consent does not mean we should be surrendering our consent to anyone and everyone.
They didn't do that. They were not actively promoting creators. it was the opposite. They were letting people mark someone as a potential recipient of contributions as a way to bootstrap their network.
> just because Chrome has the capability to coerce consent does not mean we should be surrendering our consent to anyone and everyone.
Your bias is showing.
Brave did not "coerce" anything to anyone. Their crypto stuff is opt-in. The ad blocker is opt-in.
Rationalize all you want, if you think that is justified to have a website blocking a browser like Brave because "of what they do to users", then it should be a moral imperative to help others to stop using chrome, edge and Safari.
I've seen the screenshot of the half-screen overlay pop-up advertisement that was displayed to Brave users. The "Welcome!" banner together with an actual photo of the person in question, together with the wording of the solicitation, is something that would absolutely give many, if not most, uninformed users the impression that the solicitation originated from the person featured.
Neither I, nor the linked issue, cite that Brave was blocked "because of what they do to users". If this had happened to me, I would block them based on what they did to me. As I said, the act in question is personally offensive in a way that what Google does is not. It plays on the border of identity fraud. If a browser is using my identity to solicit donations, I'm well within my right to do what I can to interfere with that.
Regards to coercion, I did not say that Brave coerced anyone. I pointed out that Google effectively does via its monopoly power, and that is why that people cannot realistically choose to block Chrome. The matter of coercion is addressing your complaint that they aren't also blocking Chrome, not a criticism of Brave.
> the act in question is personally offensive in a way that what Google does is not.
A perceived, harmless, unintentional and nonetheless remediated offense is worse than the continuous abuse of power and anti-user practices from Google, Microsoft and Apple. It might seem justified to you, but to me it's just displaced indignation and illustrates why we will forever live in this corporate dystopia.
My moral compass does not change based on who is being accused, but context is fundamental to make a proper judgment.
It is hard to come up with a situation where Google would be doing these types of tricks, because Google is already the dominant player in the market and they don't want to create products that cannibalize their own revenue streams.
I think this is about being mad at Brendan Eich, the current CEO of the company behind Brave, for his opposition to legal gay marriage in the late 2000s/early 2010s. A lot of Lobsters moderators are queer and/or politically sympathetic to queer activism.
Which goes to show the importance of judging people by their actions and not their opinions: are they going to boycott Apple as well, since Tim Cook gave millions to Trump?
Brendan Eich donated money to the campaign in favor of Prop 8, the 2008 California ballot proposition that banned gay marriage and that was overturned by the courts some time later. He didn't publicize this himself IIRC but the donations were public information and became well-known when he was (briefly) appointed CEO of Mozilla.
Yeah, I am not interested in playing this tape again. The actions from Brendan as an individual are completely separate from his actions at Mozilla. Mozilla did not change any policy during his tenure and Brave is not accused of any discrimination practices or hostile to any minority group.
I'm guessing you are talking with someone who is used to life in the North American suburbs, where kids need to be driven around and most of the options for activities are indoors.
Sadly, yes. The nearest park is 5 miles from me or the mall. The buses run on the hour and will get you within 2 miles of the park. They stop running around 7pm.
I don't get this logic. Putting aside that to get 33 different models you would come up with 5-6 different form factors, each of them on a distinct point in the tradeoff scale, why do you think that something is only worth doing if it can be put on an uniform supply-demand curve?
Apple made nearly $190 billion last year selling just iPhones.
If you think it costs more than $5 billion to design a phone and set up a production line, you are wildly off base. That’s the kind of money companies spend to build silicon fabs or release half a dozen new car models, not consumer products made by a contract manufacturer.
...you do know that Apple produces its own silicon, and probably uses about an entire TSMC fab's worth of capacity? In the end, the money to build that fab is coming from Apple.
Apple isn't making average consumer products with average contract manufacturers.
If there is one thing I believe that will come out of the whole AI bubble, is this: we will interact with applications though natural language and the "graphical UI" will be mostly about visual feedback of what operation is happening. The challenge for open source projects will be to define whether they will assume that their application will rely on some external LLM that will (hopefully) be good enough or if they will have to distill/train a specific model that can be smart enough to interact with the application tools.
What value is a friendship or connection that can not withstand a few months of "for the next months I will ride with some slower group" or "I may join you for the beginning but I can not go all the way"?
What does "can not" even mean in this context? You could refuse to use an e-bike just like you could refuse to use an expensive bicycle.
Perhaps more interesting, a lot of people I know do train but don't push themselves to the limit. There is always something left in the tank. If their goal is to train to improve it's very [lets say] expensive. They invest a lot for small returns.
Now what if you could keep going if you would otherwise feel the need to quit?
I cycle for a good while then have to guess if ill still be able make it back home. Usually I bet on the safe side. When I bet to low I could add a few laps around the block but this requires an odd kind of discipline that I seem to lack.
Rather than grow it turns into a maintenance routine. If I wanted to do maintenance I would do much less and less frequent.
> You could refuse to use an e-bike just like you could refuse to use an expensive bicycle.
I don't know if we are talking about the same thing. I am taking the argument from OP about "people recovering from surgery might need electric assist to keep up with the A-group", and I'm questioning this need.
If someone wants to have some assistance, then by all means go ahead and use it. I'm just not seeing why someone would need to have this sort of assistance, unless they are just doing some poor post-hoc rationalization for their wants.
"Need in order to" is a common construction in English. It refers to a necessary condition, without placing any absolute constraints on the necessity. (Who's to say how the absolute "need" is defined? Do we need to live? Do we need good food, positive social interactions, exercise, shelter?)
> "Need in order to" is a common construction in English.
I was paraphrashing. OP's original comment said "For some guys a high-end electric assist bike is how they stay riding with the A-group while they're recovering."
What I question here is, simply put, why "stay riding with the A-group" is in any way important? While one is recovering from surgery, what is so bad about riding with a B-group? Or why not ride with that A-group, but for a shorter distance?
What a weird take. It's not a case of "withstanding" a friendship, it's a case of enjoying some time with your buddies. If you enjoy spending time with the people you usually ride with why wouldn't you if given the ability to do so?
Horses were not cheap nor universally available. And cars had the obvious benefit that they did not leave literal horse shit around the city.
This "revolutionary design" does not offer any significant advantage over the existing systems for e-bikes. A regular e-bike without power is a just a regular bike. You can adapt a regular bike into an e-bike for < $600. Any run-of-the-mill mechanic can figure out how to work on a basic bike. This one will probably require some "certified Rivian expert" to work on it.
Only irrational neomania can justify being interested in this "revolution".
Tool use can be a sign of intelligence, but "being able to use a tool to solve a problem" is not the same as "being intelligent enough to solve a specific class of problems".
reply