- AMD AI director says Claude Code lost performance in February 2026 update
- Claude Code “cannot be trusted” on complex tasks per thousands of coding sessions
- Anthropic says it reduced stakes to medium, but Teams and Enterprises may go high
AMD AI director Stella Laurenzo has claimed that the Claude Code has become less efficient since around February 2026, claiming that it “cannot be trusted to perform complex technical tasks.”
Laurenzo’s criticism is not unfounded, based on the company’s analysis of over 6,800 coding sessions, nearly 235,000 tool calls, and nearly 18,000 reasoning blocks.
“Every senior engineer on my team has reported similar experiences/anecdotes,” Laurenzo wrote, noting that stop-hook violations (where Claude gave up early, shirked responsibility, or asked for unnecessary permissions) rose from zero in early March to about 10 per day afterward.
The article continues below
Claude Code is getting worse, warns AMD head
In a GitHub post, user stellaraccident (aka Stellar Laurenzo) identified a strong correlation between the introduction of thought editing (redact-thinking-2026-02-12) and a drop in performance on complex tasks. The AMD executive argues that extended reasoning can be “load-bearing” for advanced engineering.
Laurenzo also observed a shift from research-first to edit-first behavior, generating lower quality code, poorer adherence to conventions, and generally reduced reliability for long sessions.
Anthropic has already responded to the research with a multifaceted explanation. Claude Code’s Boris explained that the redact-thinking-2026-02-12 setting only hides reasoning from the UI and doesn’t actually reduce reasoning.
The company also introduced adaptive thinking with Opus 4.6, where the model dynamically decided how long to think to improve performance and efficiency.
“Some people want the model to think longer, even if it takes more time and tokens,” Boris added. “To improve intelligence more, set Effort=high via `/effort` or in your settings.json.”
With medium stake or stake=85 now the default for users, Anthropic has promised to test higher stakes for Teams and Enterprise users to “take advantage of extended thinking, even if it comes at the cost of additional tokens and latency.”
“I appreciate the depth of thought and care that went into this,” Boris also noted, crediting AMD’s Laurenzo for the analysis.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds. Be sure to click the Follow button!
And of course you can too follow TechRadar on TikTok for news, reviews, video unboxings, and get regular updates from us on WhatsApp also.
