M1 gets most of its performance-per-watt efficiency by running much farther down the voltage curve than Intel or AMD usually tune their silicon for, and having a really wide core design to take advantage of the extra instruction-level parallelism that can be extracted from the ARM instruction set relative to x86. It’s a great design, but the relatively minor gains from M1 to M2 suggest that there’s not that much more in terms of optimization available in the architecture, and the x86 manufacturers have been able to close a big chunk of the gap in their own subsequent products by increasing their own IPC with things like extra cache and better branch prediction, while also ramping down power targets to put their competing thin-and-light laptop parts in better parts of the power curve, where they’re not hitting diminishing performance returns.
The really dismal truth of the matter is that semiconductor fabrication is reaching a point of maturity in its development, and there aren’t any more huge gains to be made in transistor density in silicon. ASML is pouring in Herculean effort to reduce feature sizes at a much lower rate than in years past, and each step forward increases cost and complexity by eyewatering amounts. We’re reaching the physical limits of silicon now, and if there’s going to be another big, sustained leap forward in performance, efficient, or density, it’s probably going to have to come in the form of a new semiconductor material with more advantageous quantum behavior.
as a counterpoint, when the use-case for the tool is specifically “I want a picture that looks like it was painted by Greg Rutkowski, but I don’t want to pay Greg Rutkowski to paint it for me” that sounds like the sort of scenario that copyright was specifically envisioned to protect against – and if it doesn’t protect against that, it’s arguably an oversight in need of correction. It’s in AI makers and users’ interest to proactively self-regulate on this front, because if they don’t somebody like Disney is going to wade into this at some point with expensive lobbyists, and dictate the law to their own benefit.
That said, it’s working artists like Rutkowski, or friends of mine who scrape together a living off commissioned pieces, that I am most concerned for. Fantasy art like Greg makes, or personal character portraits of the sort you find on character sheets of long-running DnD games or as avatar images on forums like this one, make up the bread and butter of many small-time artists’ work, and those commissions are the ones most endangered by the current state of the art in generative AI. It’s great for would-be patrons that the cost of commissioning a mood piece for a campaign setting or a portrait of their fursona has suddenly dropped to basically zero, but it sucks for artists that their lunch is being eaten by an AI algorithm that was trained by slurping up all their work without compensation or even credit. For as long as artists need to get paid for their work in order to live, that’s inherently anti-worker.