Claude Code Reasoning Effort Fell, and Quality Followed
Anthropic says the Claude Code reasoning effort default was lowered from high to medium in early March to reduce latency,…
-
-
LLM Failure Modes Start in the Stack, Not the Chat
LLM failure modes are easiest to understand if you stop treating them as personality flaws, “the model lied,” “the chatbot…
-
Product Tutorial Videos Now Update Themselves
Product tutorial videos used to be a small production project: record the screen, write the script, clean up the narration,…
-
Google’s 75% AI-Generated Code Claim Hides the Real Shift
Google says AI-generated code now accounts for 75% of all new code at the company, with that code still “approved…
-
Anthropic Bans Can Hit Teams, APIs, and Billing Separately
The headline sounds like satire. It isn’t. Anthropic bans are a real policy and product issue: Anthropic’s own documentation describes…
-
AI Deathbots Sell Memory, Not Holograms
AI deathbots are software services that simulate a dead person from their digital traces: photos, voice recordings, chat logs, emails,…
-
66 Tokens Make a Diffusion Language Model Look Easy
A diffusion language model generates text by starting from masked or otherwise corrupted tokens and iteratively restoring them. In this…
-
Claude Opus 4.7 Token Usage Jumped 35% Without Price Hikes
Claude Opus 4.7 token usage went up for reasons Anthropic documented in its own launch materials. The list price stayed…
-
AI Illiteracy Turns Copilot Use Into Data Leaks
A common AI illiteracy failure looks mundane: staff paste confidential text into a consumer chatbot to rewrite it in a…
