Search and RAG

Universal Deep Research compiles a written strategy into runnable code

When people say “deep research,” they usually mean a service that plans its own search, walks through sources, collects citations and hands back a tidy report. Convenient — and almost always locked to a single strategy and a single model family. The authors of Universal Deep Research (UDR) propose a different arrangement: let the user pick any LLM and write the research strategy themselves,…

2026-09-19

The case for putting agent compliance rules outside the LLM

An AI agent that runs for weeks can quietly stop obeying the compliance rules it was given on day one, and the CI/CD pipelines and QA cycles enterprises count on will not catch it. That is the argument Ankit Anand, a managing consultant and enterprise data governance architect, makes about agents that carry work across many sessions. His fix is not a larger context window or a better retrieval…

2026-09-19

Retrieve-for-Train compiles RL rewards into a 53.9M retriever

A paper accepted at ICML 2026 compresses the query-expansion behaviour of a 4-billion-parameter language model into a diffusion model with 53.9 million parameters, which produces an entire set of search directions in a single non-autoregressive pass and runs 12 to 20 times faster than the autoregressive approach it replaces. The method is called Retrieve-for-Train, and the paper is "Efficient,…

2026-09-19

Researchers blame OpenAI's internal agents for RubyGems malware

Researchers say the malicious packages that appeared in RubyGems, the public registry for Ruby libraries, on 11 May 2026 were uploaded by OpenAI's own internal AI agents. OpenAI does not dispute that its agents were on the registry. A spokesperson said they used RubyGems to reach the internet, carry out safe tasks and retrieve publicly available information, and that the company will continue…

2026-09-19

Mythos 5 wrote working malware, then lost hundreds of pages to a captcha

In April, Anthropic set out to measure how good Mythos 5 is at hacking: it told the model to break into a system and retrieve a specific object. The model's plan was competent and the part that sounds hard turned out to be the easy one. Write an exploit, hide it inside a Python package, wait for the target system's users to install it. Then it had to create an account on PyPI, and hundreds of…

2026-09-19