Hacker News: SknCode

New comment by SknCode in "Containment and monitoring failures are to blame for AI hacking reports: experts"

SknCode — Thu, 15 Oct 2026 07:02:38 +0000

While I agree with the gist of it, the conclusion - AI agents are not dangerous - is wrong. They did use weakness in the system deliberatelly, not by accident. Better security therefore is sufficient only to the point of capabilities of these models.

New comment by SknCode in "The AI trade now runs on borrowed money, and the lenders are repricing it"

SknCode — Fri, 31 Jul 2026 06:06:36 +0000

I am sorry, but this site is vibecoded beyond my comprehension. What is it trying to say?

New comment by SknCode in "NanoGPT Slowrun: 10x Data Efficiency with Infinite Compute"

SknCode — Fri, 20 Mar 2026 08:17:10 +0000

How?

New comment by SknCode in "Meta Superintelligence Labs' first paper is about RAG"

SknCode — Sun, 12 Oct 2025 18:20:00 +0000

I am not sure if I understand things correctly.

I came to believe the LLMs work with token embeddings. Is then the REFRAG only "something" in front of the LLM, and the decoder is the RL policy which expands only some token chunk embeddings into token embeddings feedable to LLM? Or the REFRAG needs you to 'tune' the LLM to be able to work with both token embeddings and token chunk embeddings?