WebFovea, a paper by Jiangang Han, reports a disquieting browser agent failure. A mismatch between screenshot and browser coordinates placed clicks at three quarters of their intended position. A click meant for a point 400 pixels across landed 300 pixels across. The mismatch came from the service that runs the model, which shrank each screenshot …

I Broke a Browser Agent on Purpose. Spotting the Fault Was Not the Same as Fixing It Read More »

We live in an interesting technological era, with so many questions about the morality and legality of using AI for creative content. As such, many researchers have taken it upon themselves to find patterns in what AI generates. They tried to put into numbers what we feel when we read or see something AI-generated. Basically, …

What Makes AI-Written Stories Feel Off? Read More »

Picture an agent that answers questions about customer invoices. It has one tool, a lookup against an internal API, and a task that needs four lookups. The first two go through fine. Then the credential behind that API expires, and every call from the third one onward comes back as a 401. What can happen …

Agentic AI in Production: Why Your Agent Keeps Answering After Its Tools Fail Read More »

Job postings for data analysts, statisticians, and researchers increasingly list “AI experience” as a requirement. But that phrase means very different things depending on who wrote the listing, what industry they’re in, and what problem they’re trying to solve. A hiring manager at a retail company, a biotech startup, and a financial services firm may …

What Companies Actually Mean When They Say They Want “AI Experience” Read More »

ContentsWhen a Straight Line Stops Being EnoughFrom Straight Lines to Flexible CurvesPolynomial Regression Through a Linear-Model LensWhen One Global Polynomial Is Not EnoughHow the B-Spline Basis Is FormedNot All Bad Data Is the SameWhen Measurements Are MissingThe Outlier ProblemThe Same Six Estimators, a Different BasisStress-TestingWhat the Stress Tests RevealRicher Models Need More DataThe Contamination Pattern …

Don’t Let Messy Data Bend Your Curves Read More »