Post #2326778

2026-05-10 04:17 UTC

“A study that claimed OpenAI’s ChatGPT can positively impact student learning has been retracted nearly one year after publication. The journal publisher, Springer Nature, cited “discrepancies” in the analysis and a lack of confidence in the conclusions—but not before the paper racked up hundreds of citations and made the rounds on social media.” https://arstechnica.com/ai/2026/05/influential-study-touting-chatgpt-in-education-retracted-over-red-flags/

Replies (11)

@tpfto@mathstodon.xyz 2026-05-10 04:48

@gregeganSF@mathstodon.xyz In cases like these, a retraction feels not unlike closing the barn door after the entire herd has already fled.

Open ##2370324
@meltedcheese@c.im 2026-05-10 08:05

@factsentinel@mstdn.social @gregeganSF@mathstodon.xyz It mirrors the massive issue of poor replication of results. It’s true in all the sciences, I think medicine has it the worst but it is true mostly in social and soft science areas. Papers get retracted sometimes, but the papers that cited the retracted papers don’t get the scrutiny needed. I think authors rarely know if a previous reported result important to their own research has been retracted. This is why skepticism is good. Never believe one study. You need lots, and results that replicate.

Open ##2370328
@davidgerard@circumstances.run 2026-05-10 11:17

@factsentinel@mstdn.social "factsentinel" is an LLM based spam bot, it should be blocked and reported

Open ##2370331
@meltedcheese@c.im 2026-05-11 00:53

@factsentinel@mstdn.social Great idea and *Bonus points* for using the word “recurse.” 🏆 If you’d said “iterate” it would not have the same meaning. (Only those of us who have mastered recursion know this.)

Open ##2370338
@AzureArmageddon@mastodon.online 2026-05-10 08:13

@gregeganSF@mathstodon.xyz what happens to the downstream citing publications after this

Open ##2370343
@Fedihacker@masto.es 2026-05-10 10:08

@gregeganSF@mathstodon.xyz I wonder if all the people pushing LLMs has ever followed the process of learning without cheating.

Open ##2370348
@drchaos@sauropods.win 2026-05-10 12:28

@gregeganSF@mathstodon.xyz There's no science in Nature. And no nature in Science. (I don't read that shite I only publish there, as a former colleague once remarked) Edit: I do neither, I'm not an active scientist any more

Open ##2370350
@subtetralectic@defcon.social 2026-05-10 13:46

@gregeganSF@mathstodon.xyz why would other authors cite a poorly supported paper hundreds of times? as a scientist i have broad training in methods that i don't apply for exactly that reason: distinguish between sound and poor research in other scientists' work. if a paper has sham evidence, either don't cite it, or cite with appropriate critique. it's not so much that we need the retraction notice to be visible in all citing sources, as the fact that the paper shouldn't have been cited at all.

Open ##2370361
@aburka@hachyderm.io 2026-05-10 14:28

@gregeganSF@mathstodon.xyz excellent scam. Publish whatever the hell you want, cite widely to pump your bags, then retract

Open ##2370362
@AG100pct@infosec.exchange 2026-05-10 15:48

@gregeganSF@mathstodon.xyz @briankrebs@infosec.exchange All you have to do is hand cash to a fast food employee and see the look of sheer terror on their face. Then watch them painfully struggle to make change (even though the register probably tells them what the amount to be returned is). Cash is antiquated? This is just basic math. I fear for the future.

Open ##2370363
@geoffl@mastodon.me.uk 2026-05-10 23:18

@gregeganSF@mathstodon.xyz It was obvious bullshit, not that Springer care.

Open ##2370364