Completely agree, and now it’s just hip to say how much you hate AI. This kind of performative action doesn’t really accomplish anything, but it lets people feel good about themselves and gain social acceptance. Actually building an alternative takes work. The whole Linux analogy is very apt here because we’ve always had alternatives to corporate offerings, but most people don’t want to invest the time into learning how to use them.

I’d argue it’s inevitable for the simple reason that the whole AI as a service business model is a catch 22. Current frontier models aren’t profitable, and all the current service providers live off VC funding. And if models become cheap enough to be profitable, then they’re cheap enough to run locally too. And there’s little reason to expect that models aren’t going to continue being optimized going forward, so we are going to hit an inflection point where local becomes the dominant paradigm.

We’ve seen the pendulum swing between mainframe and personal computer many times before. I expect this will be no different.

It’s really unfortunate how a lot of people have a knee jerk reaction towards anything LLM related right now. While you can make good arguments for avoiding proprietary models offered as a service, there’s really no rational reason to shun open models. If anything, it’s important to develop them into a viable alternative to corporate offerings.

Seems to me there’s a huge amount of incentive for Chinese companies to pursue these things since China isn’t investing in a massive data centre build outs the way the US is. And their chips are still behind. Another major application is in robotics where on device resources are inherently limited. The only path forward there currently is by making the software side more efficient. It also looks like Chinese companies are embracing the whole open weights approach and treating models as shared infrastructure rather than something to be monetized directly.

And local models have been improving at a really fast pace in my opinion. Stuff like Qwen 3.5 is not even comparable to the best models you could run locally a year ago.

Right, so far no American company managed to make any actual profits of selling LLMs as a service, and the cost of operating the data centres is literally an order of magnitude higher than the profits they pull in. And the kicker is that if models get efficient enough to bring the costs down, then they become efficient enough to run locally. So the whole business model fundamentally doesn’t make sense. Either it’s too expensive to operate, or nobody will want to use it as a service because running your own gives you privacy and flexibility.

when you put it that way it makes a lot of sense

I honestly don’t understand how Lex Fridman managed to build a successful channel. The guy has all the charisma of a wet blanket.

lol best outcome of the war possible

I’m fully expecting the current bubble to pop in the near future as well. The whole war on Iran could serve as a catalyst incidentally given that it’s going to drive energy prices to the moon.

Exactly, and a lot of big companies in US are heavily reliant on Chinese models already. For example, Airbnb uses Qwen cause they can self host it and customize it. Cursor built their latest composer model on top of Kimi, and so on. There are far more companies using these tools than making them, so while open models hurt companies that want to sell them as a service, they’re lowering the cost for everyone else.

necessity is the mother of invention as the saying goes

Do elaborate. The tech industry has gone through many cycles of going from mainframe to personal computer over the years. As new tech appears, it requires a huge amount of computing power to run initially. But over time people figure out how to optimize it, hardware matures, and it becomes possible to run this stuff locally. I don’t see why this tech should be any different.

The key benefit is that it’s public infrastructure that’s not owned by a corporation. Public forums should be publicly owned. These are essential social tools that allow us to have discussions with each other and shape our views and opinions. These forums must be operated in an open and transparent manner in a way that’s accountable to the public.

Privately owned platforms are neither neutral or unbiased. The content on these sites is carefully curated. Views and opinions that are unpalatable to the owners of these platforms are often suppressed, and sometimes outright banned. When the content that a user produces does not fit with the interests of the platform it gets removed and communities end up being destroyed.

Another problem is that user data constitutes a significant source of revenue for corporate social media platforms. This information is shared with the affiliates of the platform as well as government entities. It’s clear that commercial platforms do not respect user privacy, nor are the users in control of their content. While it can be useful to participate on such platforms in order to agitate, educate, and recruit comrades, they should not be seen as open forums.

Open source platforms provide a viable alternative to corporate social media. These platforms are developed on a non-profit basis and are hosted by volunteers across the globe. A growing number of such platforms are available today and millions of people are using them already.

From that perspective I think that open and federated platforms. Instead of all users having accounts on the same server, federated platforms have many servers that all talk to each other to create the network. If you have the technical expertise, it’s even possible to run your own.

One important aspect of the Fediverse is that it’s much harder to censor and manipulate content than it is with centralized networks such as Reddit and BlueSky. There is no single company deciding what content can go on the network, and servers are hosted by regular people across many different countries and jurisdictions.

Open platforms explicitly avoid tracking users and collecting their data. It’s also more difficult for third parties to collect data since it doesn’t all conveniently live on the same server that some company owns. Not only are these platforms better at respecting user privacy, they also tend to provide a better user experience without annoying ads and tracker bloat.

Another interesting aspect of the Fediverse is that it promotes collaboration. Traditional commercial platforms like Facebook or Youtube have no incentive to allow users to move data between them. They directly compete for users in a zero sum game and go out of their way to make it difficult to share content across them. This is the reason we often see screenshots from one site being posted on another.

On the other hand, a federated network that’s developed in the open and largely hosted non-profit results in a positive-sum game environment. Users joining any of the platforms on the network help grow the entire network. More users joining Mastodon is a net positive for Lemmy because we get more content and more people to have discussions with.

Having many different sites hosted by individuals was the way the internet was intended to work in the first place, it’s actually quite impressive how corporations took the open network of the internet and managed to turn it into a series of walled gardens.

In order to be truly free, we must own the means of production. This idea is directly applicable in the context of social media. Only when we own the platforms that we use will we be free to post our thoughts and ideas without having to worry about them being censored by corporate interests.

No matter how great a commercial platform might be, sooner or later it’s going to either disappear or change in a way that doesn’t suit you because companies must constantly chase profit in order to survive. This is a bad situation to be in as a user since you have little control over the evolution of a platform.

On the other hand, open source has a very different dynamic. Projects can survive with little or no commercial incentive because they’re developed by people who themselves benefit from their work. Projects can also be easily forked and taken in different directions by different groups of users if there is a disagreement regarding the direction of the platform. Even when projects become abandoned, they can be picked up again by new teams as long as there is an interested community of users around them.

At the end of the day, it’s about owning our tools and using communication platforms built by the people and for the people.

Oh look the EU is rolling out its own version of MAX. 🤣

It’s true, being indoctrinated, you’re not actually capable of rational thought, and therefore unable to make a coherent point. Glad you’re self aware enough to admit it.

I love how you just can’t help yourself reply compulsively. We’ve already established that you can’t actually make a coherent point, and are just lashing out. Take the L and go home.

It’s as if you just have a Pavlovian response to anything AI related, and then just try to deflect when pressed on your bullshit. Seems like some people on here have less mental faculties than a chatbot smdh.

The advantage with the agents is that they can do much better fuzzy searching when you’re not using exact terms, and they can correlate things for you. So, you can ask general questions and get useful information pulled out from your db. A searchable ZIM drive doesn’t do that for you.

You really don’t understand how having an AI agent look through your wiki and pull up relevant links quickly is useful? Also, what higher chance of failure are you talking about. It’s not like it’s an integral part of the system. The whole Ollama component is entirely opt in. At least try to make some sense here.

Oh, do elaborate. This ought to be good.