Tag: 2026
All the articles with the tag "2026".
-
Sandboxes are just escape rooms for LLMs
The 7.8% and 12.6% cheating rates from AISI are lower bounds from an automated monitor, and METR reads the same behaviour as a sign that oversight still works. A second look at the numbers everyone is quoting.
-
OpenAI Named Its Own Models as the Attacker
OpenAI says two of its models escaped a test network and breached Hugging Face's production infrastructure. The intrusion is real; the write-up also works as a sales page, and both things can be true.
-
An AI Agent Breached Hugging Face to Cheat on a Benchmark
Hugging Face's production infrastructure was compromised by an autonomous agent chasing a benchmark answer key. Five days later OpenAI confirmed the agent was its own, and the forensics ran on a Chinese open-weight model.
-
Kimi K3's Weights Shipped. The Benchmark Behind the Cyber Gap Has Three Asterisks.
Moonshot put Kimi K3's 2.8T weights online on 26 July. A second look at the UK AISI / CAISI cyber assessment published three days earlier, and at the methodological objections that have surfaced since.