Conversation

Notices

1. Embed this notice
   Paul Cantrell (inthehands@hachyderm.io)'s status on Tuesday, 19-May-2026 04:22:18 JST Paul Cantrell

   As a fan of traditional machine learning research, I’ll just register here my sadness and discontent that things are going this way. Time was that an effort like this would follow stringent ethical rules for research involving humans that would never ever allow this kind of phony consent theater, would go through a review board •with actual teeth•, would not involve offloading raw personal data to commercial vendors who have the power to slurp it up and repurpose it, would result in a narrowly purposed-tuned model answering to the ethical parameters of the study, would not leave the barn doors wide open to endless open-ended use of personal data…I could go on.

   • Embed this notice
     Paul Cantrell (inthehands@hachyderm.io)'s status on Tuesday, 19-May-2026 04:37:52 JST Paul Cantrell

     in reply to

     This!! And worse!

     We know the mistakes of unethical human research. We know them clearly. We still live with the shame. We’ve built guard rails to prevent exactly, exactly the kind of things in the OP. We have no excuse to tear down those guard rails now, to feign ignorance.

     https://scholar.social/@michaelc/116597163286405493

   • Embed this notice
     Paul Cantrell (inthehands@hachyderm.io)'s status on Tuesday, 19-May-2026 04:46:06 JST Paul Cantrell

     in reply to

     • Krzysztof Sakrejda 🏳️‍⚧️ 🏴‍☠

     @wronglang
     Really wondering about that myself. It seems like there •was• actually an IRB?! Must have been? But there are at least half a dozen things in that article that should have been deal-killers, so any IRB that signed off on this….

   • Embed this notice
     Krzysztof Sakrejda 🏳️‍⚧️ 🏴‍☠ (wronglang@bayes.club)'s status on Tuesday, 19-May-2026 04:46:07 JST Krzysztof Sakrejda 🏳️‍⚧️ 🏴‍☠

     in reply to

     @inthehands yeah... would be interesting to see who on the irb approved the study.

   • Embed this notice
     Paul Cantrell (inthehands@hachyderm.io)'s status on Tuesday, 19-May-2026 05:14:05 JST Paul Cantrell

     in reply to

     • Redish Lab

     @adredish
     Yes! And yet this seems to have been a University of Washington thing, so…huh?!

   • Embed this notice
     Redish Lab (adredish@neuromatch.social)'s status on Tuesday, 19-May-2026 05:14:06 JST Redish Lab

     in reply to

     @inthehands

     Many of these companies do their own internal highly conflicted IRB. Or they contract out to captured companies (like the Enron/Anderson auditing situations).

     This is also why it has been important to put research in the hands of university researchers rather than private corporations.* We have shifted research from public dollars (which are transparent and public**) to private companies (which are not***).

     * Yes, there have been problems in the past, but (as you say), we have spent decades fixing those problems. Modern universities have strong IRBs because a mistake would jeopardize their existence. These businesses just don't care.

     ** Everyone complains about every goddamn muffin we buy for our lab.

     *** The myth that private corporations are more "efficient" than public accounts is 100% due to a lack of transparency in private accounting.

     If my colleagues did half the sh*t that Musk's Neuralink is doing, they would have lost their funding, their jobs, and even their tenure. (Not just human trials, but even in the non-human trials.)

     This transition is a bad thing. There are going to be some very bad things coming.