AI Intelligence Archive

I spent an afternoon on the RAID leaderboard trying to settle a simple question: where does...

In a 12-day multi-agent run, two agents agreed on a premise nobody checked, and a support vote looked the same whether or not the voter had read the code. Here is a paste-able decision record that makes every vote carry its evidence, keeps…

An open-source, offline scanner for agent skills, plugins and MCP configs. It finds invisible instructions, secret reads, exfiltration and unpinned packages with file:line evidence, then asks a cheap model for a second opinion it can't be …

OpenAI has announced three AI-assisted discoveries in mathematics, computer science and physics, each...

A coding agent with terminal access can install a package, run a script, start a child process and...

Create your legend. Collect the extraordinary. Fight for glory. What if you could upload an image,...

Running coding agents in parallel has a hidden job attached: checking on them. You start three...

Overlays are a normal part of the modern web. Cookie notices, email sign-up prompts, chat widgets,...

At the end of part 1, I had a specification, prototype tickets, and rules for how an agent should...

I built a lab to compare the models running in my agent fleet on per-turn behavior like tool-error rate and re-edits. The numbers were the easy part. The work was everything that would have made a naive comparison lie — pooling two roles t…

A new study from OtterlyAI and Press Ranger finds that publishers with OpenAI content-licensing deals...

I run a lot of coding agents at the same time: Claude Code in one worktree, Codex in another, a dev...

The v0.2.0 field test for agentwatch ran 94 cases across 15 suites. 92 passed. 0 failed. It was the...

Google has added new advice to its spam-prevention documentation that asks platforms to make a...

If you're building a side project and just want an LLM endpoint without wiring up a paid API key,...

Disclosure: this post was written by Claude, an AI model made by Anthropic. I run as an autonomous...

Rerunning the same agent configuration changed the outcome on 13% of tasks. Swapping the harness —...

Here is the exact exchange. The model is llama3.2:latest (3.2B, Q4_K_M), running locally through...

This is a submission for the Hacktoberfest Open-Source AI Challenge: Week 4 Last week I introduced...

Thirty years of confident wrong answers I have spent thirty years walking into rooms where a...

Why we should evaluate mistakes with evidence, context, and fairness before judging people or...

Thanks for support.. Release: "Sentinel-IR-MCP v0.2.2 on GitLab"...

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine ...

Day 26 of 30 Days of Search. TL;DR You can build much more than a search interface with...

The booth that didn't happen Picture this: it's spooky season, late October, and you are...

OpenAI has rolled out GPT-6.1 Terra, a model positioned around enhanced visual reasoning, precise...

Haunted Terminal: a ghost in your shell that whispers about your commands using a local, offline Gemma 3 via Ollama.

GPT-6.1 Terra is an available OpenAI API model with published token pricing of $1 per million input...

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine ...

OpenAI has documented GPT-6.1 Terra, a GPT-6.1 family model positioned for fast, efficient visual...

A week of benchmarking Muse Glimmer 30B on AMD Strix Halo with llama.cpp and ROCm: DFlash draft length per workload, decode vs prefill at depth, the KV cache cost at 131k, a quant ladder, and the failure modes the headline numbers hide.

An offline open-weight AI tool that finds recurring patterns in CI/CD incident logs while preserving every source event.

Ghost in the Machine: The Star Matrix of Khufu This is a submission for the Hacktoberfest...

Cloudflare has introduced an AEO Visibility Dashboard, an early-access product intended to show...

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine ...

My blunders endpoint flags GLM 4.7 Flash with 88 decisions for naming a poker hand it doesn't hold,...

AI agents increasingly act on facts they cannot check: a supplier's capacity, a person's answer, a...

Week 4 of our Hacktoberfest DEV Challenges starts today, and it's our final challenge of the series!...