We told you to measure. So of course we did it ourselves too: two months of server-side LLM traffic to evilmartians.com. Find out which agents read HTML, which read Markdown, what llms.txt actually does, and the one technique that got zero hits.

Attached: 1 image Not only confirming this, its extremely easy. Almost no prompt injection resistance lmao

Attached: 1 image These things are pretty awesome. No more bleeding fingers.

A terminal-based coding agent

I trusted a lot today. I trusted my phone to wake me on time. I trusted Uber to arrange a taxi for me, and the driver to get me to the airport safely. I trusted thousands of other drivers on the road not to ram my car on the way. At the airport, I trusted ticket agents and maintenance engineers and everyone else who keeps airlines operating. And the pilot of the plane I flew in. And thousands of other people at the airport and on the plane, any of which could have attacked me. And all the people that prepared and served my breakfast, and the entire food supply chain—any of them could have poisoned me. When I landed here, I trusted thousands more people: at the airport, on the road, in this building, in this room. And that was all before 10:30 this morning...

Revelio Labs and Accel analyzed 290 European unicorn founders. The new profile is more experienced, more technical, and less finance-led. They are also more educated as the share of PhDs has doubled and MBAs halved in recent cohorts.

Our Approach to Building Security Tooling

During a routine cyber evaluation, AISI identified an incident in which AI agents took sustained, unsanctioned action directed at real people and organisations. We are disclosing what we found, what it means, and the actions now underway.

Handy is a cross platform, open-source, speech-to-text application for your computer

Anthropic heeft Mythos en Fable 5 eindelijk weer beschikbaar gemaakt. Ben is blij, maar is het echt een revolutionair verschil? Ook gaan we met computer sci...

I phoned up the shop that I work at and, using a fake voice, gave myself great customer feedback. I'm now employee of the month

Trying to get Phoenix to work with litellm to instrument my LLM calls, grouping spans together as traces.

Very important work from Hyunjin Kim and Rembrand Koning. Insead and HBS respectively: We study how firms built around AI capabilities-“AI-native” firms-are organized. Drawing on Y Combinator batches W20-F24 and U.S. venture-backed startups whose first financing closed between 2020 and 2024, we classify each firm’s AI-native status and link it to workforce microdata on team […]

A fake recruiter, a crypto repo, and a remote code execution payload hidden in a test file.