AI tools have lowered the cost of writing code to near zero. What they haven't changed is which decisions require human judgment — and for builders without engineering backgrounds, the risk is not moving too slowly. It's not knowing which decisions to own.
Photo: Unsplash
Researchers at Berkeley found that eight of the most widely cited AI agent benchmarks — including SWE-bench Verified and OSWorld — can each be exploited to achieve near-perfect scores without solving a single task.
A developer analysed six months of Claude API session logs and found Anthropic silently shifted the default prompt cache TTL from one hour to five minutes on March 6, causing measurable cost increases for long-running sessions.
SQLite 3.53.0 adds ALTER TABLE support for adding and removing NOT NULL and CHECK constraints, closes a long-standing gap that previously required workarounds, and ships a new JSON array insert function along with CLI improvements.
AISLE tested eight models against the vulnerabilities Anthropic's Mythos found, and the results undercut the frontier model exclusivity argument: a 3.6B parameter model at $0.11 per million tokens found the same FreeBSD zero-day.
The Linux kernel's official documentation now defines how AI coding assistants should be attributed, establishes that AI agents cannot sign off contributions, and places full legal responsibility on the human submitter.
As training costs reach billions of dollars, fewer organisations will sustain frontier-level open releases — Nathan Lambert argues a collectively-funded consortium is the only viable long-term mechanism.
GitHub Copilot and the Vercel Claude Code plugin both defaulted to collecting developer data in the same week. The exemptions reveal the business model: enterprise users pay in cash, everyone else pays in training data.
AI assistants recommend packages they know, not the simplest solution. For a new generation of builders, that default is creating dependency footprints that more experienced developers would never accumulate.
Retrieval-augmented generation solves a knowledge access problem. Most teams reaching for it are trying to solve an output reliability problem. Those are different problems with different solutions, and confusing them is expensive.