You have used chatgpt presumably. Based on your interactions with it, do you seriously think it should be allowed to shoot a gun without any human oversight?
AFAIK the signal backups use symmetric encryption with user generated and controlled keys and anonymous credentials (https://signal.org/blog/introducing-secure-backups/). Do you have a link about the usage of sgx there?
Also fwiw I think tees and remote attestation are a pretty pragmatic solution here that meaningfully improves on the current state of the art for llm inference and I'm happy to see it.
Stagex is a remarkable achievement and one of the most exciting projects that I have encountered this year. I plan on migrating a few high value build pipelines in the near future. Thank you for the excellent work.
With that said, I also write a lot of Haskell and would be very sad if nixos dropped support because it was not yet fully bootstrappable. The NixOS supply chain and build pipeline could absolutely be meaningfully hardened, but I think that given the state of the ecosystem at large, and the project's widespread usage as a general purpose OS, achieving the kind of trust model and security guarantees offered by something like stagex is not yet realistic without making usability compromises that most of it's userbase would not find acceptable.
If you're not making changes to the bootloader it's essentially impossible to brick nixos: updates are fully atomic and every change can be rolled back by booting into an old generation.
This combined with the fact that the full source code for the system is contained within a single monorepo that I can checkout and grep through makes NixOS the easiest to understand and most transparent distro I have ever used.
maybe something to do with Haskell being a beautiful and foundational language that has been a major driver of progress in programming language design for the last two decades?
Having seen LLMs so many times produce coherent, sensible and valid chains of reasoning to diagnose issues and bugs in software I work on, I am at this point in absolutely no doubt that they are thinking.
Consciousness or self awareness is of course a different question, and ones whose answer seems less clear right now.
Knee jerk dismissing the evidence in front of your eyes because you find it unbelievable that we can achieve true reasoning via scaled matrix multiplication is understandable, but also betrays a lack of imagination and flexibility of thought. The world is full of bizarre wonders and this is just one more to add to the list.
I do not think this is niche in the slightest. I would very happily take a 2-4x slowdown for almost all of the web facing C software I run if I get guaranteed memory safety. I will be using at the very least fil-c openssh (and likely much more) on every machine I run.
128 is still not 300. Something like 4x 6000 blackwell is the minimum to run any model that is going to feel anything like claude locally.
To my deep disappointment the economics are simply not there at the moment. Openrouter using only providers with zero data retention policies is probably the best option right now if you care about openness, privacy and vendor lock-in.
The hardware required to run something like deepseek / kimi / glm locally at any speed fast enough for coding is probably around $50,000. You need hundreds of gigabytes of fast VRAM to run models that can come anywhere close to openai or anthropic.