samhita-alla

Supply chain attacks are a reasonable use case. But do they require VM-level isolation?

For example, what about a sandbox that isolates arbitrary user code in its own process using kernel-level primitives like user namespaces, Landlock and seccomp? That's the same class of problem Claude Code and Codex are solving when they execute model-generated commands, and it should also limit filesystem and network access as you mentioned. It's process-level isolation rather than VM-level isolation, but it still provides meaningful containment. What do you think?

My concern is that people often default to VM-level isolation without really thinking through what it's meant to protect against and whether they actually need it. It comes with a fairly heavy infra burden and isn't necessarily the right answer for every use case.

OpenAI hasn't confirmed it yet, so as of now, there are no reliable sources to rely on for this information. I just removed that part.

We have many existing sources on LLMs. I referenced a couple of them that I find great. Repeating the same content doesn't offer much value. :)

And I wrote the entire content myself.

I wanted to talk about what powers LLMs, which I believe is important. The answer to that is transformers. While I may not have delved deeper into how a transformer actually works, I tried to explain the concepts in the simplest way possible.