AI slop detection still works today, but researcher fears it won't by 2027
paulnovosad · x · 2026-09-26
Responding to claims that AI writing will soon be undetectable, economist paulnovosad argues we should exploit current detection tools while they work, even if the 'post-detection' era arrives around 2027. A concise take on the shrinking window for filtering AI slop.
More from AGI Musings
- Yoshua Bengio tells UN the AI race is 'a product of choices', labs can slow down — DavidSKrueger · 2026-09-26
- New poll: only ~1 in 15 see no AI risk; majorities back a pause on both sides of Atlantic — DavidSKrueger · 2026-09-26
- Musk lays out his 90%-likely AI future: personal robots and universal high income — XFreeze · 2026-09-26
- Rude-by-default messages to AI support may kill human customer service entirely — GabGarrett · 2026-09-26
- Is AI a 'Normal Technology'? Reddit Debates the Framing — Yaoel · 2026-09-26
- Carnegie: China Passes US as Top AI Talent Hub, 40.6% to 34.2% — yogthos · 2026-09-26