dawg91

4 ms·

1.
by dawg91 7mo ago
Your eastern european users will have some interesting results when googling for this
2.
by dawg91 7mo ago
https://docs.near.ai/cloud/verification/
3.
by dawg91 7mo ago
I mean it is described somewhat succinctly no? Potentially untrusted tools are isolated from the rest of the system - there were recently some cases of skills for openclaw being used as vectors for malware. This minimizes the adverse effect
4.
by dawg91 7mo ago
Or you design the sandbox so smartly that is seamless...
5.
by dawg91 7mo ago
Yes exactly, keys are only injected at host boundary
6.
by dawg91 7mo ago
Idk this seems to be gaining momentum and with devs being able to leverage their skillset via vibe coding anything seems possible really.
7.
by dawg91 7mo ago
They do verifiable inference on TEEs for the open source models. The anthropic ones I think they basically proxy for you (also via trusted TEE) so that it cant be tied to you. VPN for LLM inference so to speak.
8.
by dawg91 7mo ago
I think the guys who are developing this (Illia Polosoukhin of "Attention is all you need") and others knows enough to leverage their skills with AI vs. producing slop
9.
by dawg91 7mo ago
Fun fact: it's being developed by one of the authors of "Attention is all you need"
10.
by dawg91 7mo ago
It's a hardened, security-first implementation. WASM runtime specifically is for isolating tool sandboxes
11.
IronClaw: a Rust-based clawd that runs tools in isolated WASM sandboxes(github.com)
149 points by dawg91 7mo ago|69 comments