Firehose

Filtered to tagged “password managers” · clear filters

All PeopleCompaniesPapersPodcastsHacker News

Browse by tag

15 SEP 2026 · Hacker News · 37 pts · 9 comments ↗

1Password's AI patching benchmark incorrectly reported that models produced clean fixes only 26% of the time, which is misleading due to four methodological flaws: (1) complex bug fixes, (2) deliberately bad instructions, (3) trials that prohibited testing, and (4) a flawed grading system. AI summary