The point was that their business model can’t sustain growth of the userbase which makes it not like any other platform. The flaw is in comparing LLM companies to other tech companies where the larger the userbase, the more profitable they become.
The point was that their business model can’t sustain growth of the userbase which makes it not like any other platform. The flaw is in comparing LLM companies to other tech companies where the larger the userbase, the more profitable they become.
I don’t understand why people ask this. Most people you talk to on Lemmy will say they don’t want the userbase to grow much more than it has because with that growth comes the other problems that larger platforms like shitter and reddit have.
That’s true by and large and we also don’t have enough moderators here as is.
And for reasons I don’t understand, people keep asking why mainstream media outlets, influencers, and other trusted accounts don’t transition to the fediverse, as if they won’t bring with them an influx of users (at least a fraction of which would be considered undesirable).
Why do you want them to come here? (As someone who would like to see Lemmy grow, I’m curious about how you think this will rollout and what the consequences will be). I would like to see Lemmy grow but I’m not sure all of that growth will have solely good follow-on effects.
Or they just didn’t know how. Which is a distinct possibility. Some devices these days even let you share the Wi-Fi password through QR code or similar. So you don’t have to enter it into every device.
I don’t understand why they’re considering a ban. People should be changing the default password on their router. If they aren’t and they leak information that isn’t theirs, tough shit, fine them. If they leak their own information, let them deal with the consequences.
Yes, but the article goes on to say that when the userbase of paid customers grows the model becomes less sustainable as those paid numbers go up. It costs more money to support new paid users than those users can pay. More compute power, more resources, and larger models (which also become exponentially expensive and unsustainable due to the lack of clean training data).