The fact that the tech changes all the time sounds like a major obstacle. I think a big reason why so few people refine their LLM usage the way Janus does is that any skill you’ll acquire through practice might become irrelevant in the next few months, when the company releases the next update. Likewise, you could probably design some nice thick interesting AI-based musical instruments right now, but the threat of immediate obsolescence would discourage people from actually practicing. Instead, we get Suno, because that’s the only thing that is viable on these timescales.
So, unless AI progress slows down dramatically, making thick AI tools will require figuring out a way to make things plausibly future-proof. That sounds hard, but maybe there’s a way.
I agree that the speed of change makes it harder to develop deep practice, but I don’t think it’s a such a big blocker.
There are many instances of practices that need to generalize and evolve as the underlying tech changes: in cyber sec, every new tech is an opportunity for breaches, so the practice of the hacker/safety mindset needs to translate fast to every new development, in programming it is common for professionals to learn new languages / frameworks every 2-4 years.
Even if the models change every few months, I think it’s fair to say we could imagine deep practice that don’t depend on the specific details of a model but can translate to new models (more like the hacker mindset of cyber sec than the muscle memory of violin).
I also think it is possible to develop interfaces (like comfyUI) that creates a layer of practice that is independent of the model changes. For instance, you can learn workflow design patterns in how to combine LoRAs, and other adaptors that can generalize across models. Though it’s probably limited currently as more advanced image generation model might make the whole LoRAs ecosystem obsolete.
The fact that the tech changes all the time sounds like a major obstacle. I think a big reason why so few people refine their LLM usage the way Janus does is that any skill you’ll acquire through practice might become irrelevant in the next few months, when the company releases the next update. Likewise, you could probably design some nice thick interesting AI-based musical instruments right now, but the threat of immediate obsolescence would discourage people from actually practicing. Instead, we get Suno, because that’s the only thing that is viable on these timescales. So, unless AI progress slows down dramatically, making thick AI tools will require figuring out a way to make things plausibly future-proof. That sounds hard, but maybe there’s a way.
I agree that the speed of change makes it harder to develop deep practice, but I don’t think it’s a such a big blocker.
There are many instances of practices that need to generalize and evolve as the underlying tech changes: in cyber sec, every new tech is an opportunity for breaches, so the practice of the hacker/safety mindset needs to translate fast to every new development, in programming it is common for professionals to learn new languages / frameworks every 2-4 years.
Even if the models change every few months, I think it’s fair to say we could imagine deep practice that don’t depend on the specific details of a model but can translate to new models (more like the hacker mindset of cyber sec than the muscle memory of violin).
I also think it is possible to develop interfaces (like comfyUI) that creates a layer of practice that is independent of the model changes. For instance, you can learn workflow design patterns in how to combine LoRAs, and other adaptors that can generalize across models. Though it’s probably limited currently as more advanced image generation model might make the whole LoRAs ecosystem obsolete.