AI Intelligence Archive

The box that earns this site its "I tested" lines died Wednesday and stayed dead three days. What shipped anyway, what refused to, and day four's clean answer.

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle ...

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle ...

Symantec says Aoki now steals Claude, Codex, Cursor and Copilot credentials. Find the tokens your agent configs hold in plain text, and what to rotate.

Anthropic's agents exploited real websites this month. A practical guide to sandboxing AI agents: allowlists, egress control, kill switches and audit logs.

You already use prompting every day. You just don't know the names of what you're doing. Every time...

The alarm goes off at 3am. By the morning, three people can tell you three different things about...

A commercial cleaning company asked me to look at their AI receptionist because "it keeps giving the...

A small HNSW index in TypeScript, 50,000 random vectors, and a script that measures recall@10 against brute force at efSearch 10, 40, 100 and 200.

Labs have documented models special-casing and editing tests to pass. Treat agent-written tests as claims, protect a suite, flag turns that edit both.

I've been spending a lot of time thinking about where all the recent advancements in AI are going to...

Written with AI assistance; every number comes from the war records. I build SECOND STRIKE, a...

OpenAI’s Dots are persistent AI agents designed to operate with dedicated cloud compute, memory, and...

Pull every unresolved review thread with gh api graphql, build one prompt with jq, run claude -p or codex exec, then check each commented line changed.

One ALTER TABLE RENAME COLUMN breaks a rolling deploy. Run the safe Postgres version: add, sync with a trigger, backfill in batches, switch, drop.

Google AI Overviews can send meaningful referral traffic, but standard analytics reports may not show...

Getting an AI app to answer a question is exciting. Getting it to recognize when the documents don’t...

A temporary change in how ChatGPT handled outbound links distorted a key marketing metric during the...

Before Claude Code 2.1.299, an allow rule for mcp__a also allowed servers named a__b and a_. Why the names collide, a test, and a grep for your settings.

https://developers.openai.com/blog/maintaining-software-with-the-gpt-6-family

Bartender R2D66 explained the next step in evolution. Whisky made it sound reasonable. Welcome to...

This is a submission for the Hacktoberfest Open-Source AI Challenge, Week 2: Side Hustle ...

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle ...

Every fresh agent session sends a system prompt, tools, MCP and memory files first. Measure it in Claude Code, Codex, Copilot CLI and OpenCode.

Building an agent is easy now. Trusting it is the hard part. Here's how I took one agent all the way,...

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle. ...

I wanted a local model for the privacy it gives me. And if I do have to use a frontier model as the...

Posit has made disclosure of material AI involvement an explicit expectation in its Posit AI Pass...

https://app.lumalabs.ai/

Pixel-space diffusion models have recently re-emerged as a compelling alternative to latent diffusion, avoiding the lossy latent bottleneck and separately trained autoencoder. JiT shows that a vanilla diffusion transformer can generate dir…

Fine-tuning large language models (LLMs) can induce behaviors beyond those explicitly targeted. For example, training a model to respond more warmly can make it more sycophantic, and training it to write insecure code can cause broad misal…

Reproducible and analytically robust research findings are essential to scientific progress, but evaluating these properties typically requires costly manual effort. We investigate whether LLM agents can help automate reproduction and robu…

Scientific discovery requires developing worthwhile hypotheses into practical investigations, yet learning from this evolving research process remains challenging. We introduce an operational definition of valuable and testable research hy…

Parameterized quantum circuits are widely investigated in quantum machine learning, yet how their architecture can provide task-aligned inductive biases remains poorly understood. We study data reuploading quantum circuits with two distinc…

A reasoning trace can accompany a correct answer without providing effective supervision for a particular student. We study how to construct traces that improve learning beyond answer-only supervision by considering both their intermediate…

A planner with a fixed budget of preventive help must decide whom to help and when. Waiting reveals risk but removes intervention opportunities as higher-risk individuals leave first. We characterize this tradeoff under repeated noisy sign…

Future quantum networks will rely on control-plane decisions such as routing, scheduling, entanglement provisioning, purification, and memory allocation. Each decision depends on a timely and reliable estimate of the quality of the entangl…

Self-driving laboratories (SDLs) accelerate scientific discovery by automating experimental execution and optimization. AI agents can increasingly propose new experiments, but existing SDLs can only run the procedures they were built for. …

The evidence, $\mathcal{Z} = p(x | \mathcal{M})$, of a data-generating model $\mathcal{M}$ is a powerful tool in Bayesian frameworks: it enables model inference, model misspecification tests, and goodness-of-fit assessment as an alternativ…

Many of the most important datasets in the physical sciences live natively on the sphere, and these data often carry no preferred position or orientation. Using methods that do not reflect this symmetry leads to far higher simulation requi…