Anthropic's Claude Opus 5 Criticized as AI Demo Over Practical Use

Industry observers note that flashy AI demonstrations are eclipsing tools that actually improve work efficiency. Anthropic's Claude Opus 5, while praised for generating impressive 3‑D game scenes, is described by power users as verbose, scattered and difficult to steer for knowledge‑work tasks, with control issues that limit its practical value.

Other AI developments receiving attention include Retool's structured testing lab for documentation, Meta Platforms' AI agents that integrate with calendar and email applications, and Jack Dorsey's open‑source "Buzz" platform that enables collaborative model iteration. These utilities aim to turn messy workflows into repeatable systems rather than showcase visual spectacles.

At the DIA 2026 conference, regulators highlighted that AI has not yet reduced overall time spent on regulatory submissions. The FDA, EMA, MHRA and PMDA are deploying AI to extract and summarize data, but reviewers still spend significant effort locating and verifying information. The consensus is that AI must accelerate the entire submission timeline—not just draft text—to meaningfully shorten critical paths.