Claude-written code expected to be strictly better than human-written code within the year
“Claude-written code was somewhat worse than human-written code at Anthropic in late 2025, is roughly at parity today, and we expect it to be strictly better within the year.”
Context: Same Anthropic Institute essay "When AI builds itself". "Within the year" from a ~June 2026 publication is ambiguous (calendar 2026 vs 12 months).
How we judge it
Our interpretation: by 5 Jun 2027 (12 months from date_said), Anthropic or a credible independent evaluator reports that Claude-authored production code at Anthropic (or equivalent) is strictly better than typical human-authored code on Anthropic's own stated quality criteria (correctness + maintainability).
Resolution criteria are our interpretation of the claim, written before grading. Methodology
- Said
- Resolve by
- Due in 237 days
- Verdict
- Open
- Stated probability
- None stated
- Specificity
- 1 of 3 (vague or hedged)
- What changes for
- Software
Share this scorecard
Post on X LinkedIn Share image (PNG)
Embed / Cite
The embedded card is live: it shows the current verdict, with the date the data was read. If a verdict is corrected later, the card changes too, so cite the page and its date for a fixed record.