I am very excited for a future where all software is by default modifiable, even shipped binaries, via patches or trampolining, or trickery I don't even know the name of.
I actually have a panel that I've collected over time, starting with a former coworker who was visually impaired, and I am now thinking it should be some kind of users-signup-and-give-feedback system, some sort of statistical sampling of a few people across the accessibility needs spectrum.
Thus demonstrating that the final accessibility gate must be "the mark I human eyeball". Looks impossible to read without reader mode on my screen.
Your "paper" choice is #bfbfc4 - even going to #ddd makes the APCA go from ~63 to 80 and would probably address the complaints handily. WCAG has a lot of gaps.
The cool thing about cryptographic security is that it's an asymmetric amount of cost for both parties - to encode is cheap, to break is many orders of magnitude harder, unlike obfuscation where the difficulty is approximately symmetrical.
Just to clarify, latency is one form of performance, and a separate thing to optimize from total resource usage in more classic "performance critical" situations. That performance might be energy, space, or other dimensions besides latency. It might also be something like reliability, accuracy, precision, or even the very human factors like simplicity, modifiability, and visibility.
Heck, even latency alone you can just reduce the standard deviation and get smoother flows. Little's law is a great callout here too, one of my favorite computer science principles.
To concur it's "latency critical", not "performance critical", people often confuse those two - optimize it all, but especially the chained critical path latency!
I suspect that, in aggregate, all of the informational output of humanity prior to 2020 has taken more resources to produce than the last few years of LLM research.
This kind of god-of-the-gaps thing is kind of amusing to watch, especially since I can clearly see myself doing it over time.
Even if the models never get better than they are now, which is unlikely, they're already equipped to usurp taste, ethics, and everything else to a significant degree. Human feedback will always be valuable because we assign value to human feedback, but if you don't think an ethics-bot will ever be a thing, you're not seeing the current state of things.
Perhaps even more importantly, the current frontier LLM models are self-admittedly the product of enormous quantities of copyright infringement and even less savory inputs, so calling them out for distilling the fruit of that tainted tree reads as highly hypocritical at best.
The issue is when users interact with claude, what can they see? If the AI can see super-secret-payroll-details.md - is it allowed to discuss that with any user? is there some sort of internal ACL within claude that asks pretty please don't tell everyone they're underpaid?
That's why multiuser agents can be an issue.
The other way to do this, is to have claude installed for everyone, and any user's chat with claude uses their personal permissions via something like oauth delegation or AWS IAM role kinda things. But that, again, is complicated ACL logic living through the delegation of roles and access
I struggle to reconcile this attitude with, among other things, the massive number of exploits being discovered and the pretty straightforward utility for coding. Have you used codex recently?
The real challenge is reigning in the size and complexity of the codebase if you're using AI to generate it. I have a bunch of skills (YAGNI / KISS inspired) but it still requires a significant amount of human effort. Knocked down about 9000 lines of cruft this month and I know there's another 10k in there sitting around.
This is a good/fast/cheap tradeoff, but in the form of accurate/precise/creative, or something in that ballpark anyway.
You can constrain the AI to be more accurate by having it a) equivocate (lose precision) or b) turning the temperature to effectively zero, thus losing "intelligence" and creativity.
There's no such thing as a free lunch. The models perform best out of the box, and everything you add into context makes them perform worse generally, but more likely to do what you want. Every token is a burden, but without those constraints the generation won't do anything.
So by that standard, hallucination is something completely core - you can't get rid of it without losing what makes them useful. You can make different tradeoffs, or use a very smart model in a way that is completely locked down (or adversarially verified, same result) which results in far more tokens used than if you just let the model do it's thing and validated for hallucination independently, outside of the generation loop.
PDX/OR/USA
justin dot george a@t gmail.com
"Full stack JS" since 2001