Early 2025. I imagine the results would be quite different with mid 2025 models ... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		afro88 7 months ago \| parent \| context \| favorite \| on: Measuring the impact of AI on experienced open-sou... Early 2025. I imagine the results would be quite different with mid 2025 models and tools.

phyzome 7 months ago [–]

If they used mid 2025 models and tools, the paper would have come out in late 2025, and you would have had the same complaint.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact