Italian User Claims Claude Is Overhyped

Significant-Paper-40 · reddit · 2026-07-12

An engineering student shared their negative testing experience with Claude. When given a structural mechanics problem, Claude took a long time to reason through it but still delivered many inaccurate or completely off-track conclusions.

The user then tested it in a coding scenario, finding that Claude got stuck on a backend app issue they had written, while Codex solved it in about a minute with just two prompts. The author therefore questions whether the hype around Claude is overblown, even speculating that Anthropic might be exaggerating the model's capabilities for commercial, financial, or IT scenarios, and asks the community if anyone has had similar experiences.

Original post →

More from Fun

Fun channel →