This flew under the radar for some reason and needs more attention.
Over a week ago Turnitin admitted that their AI detection tool is much more prone to false positives in real-world usage than they originally said it would be. 4x higher.
And that "As a result of this additional testing, we’ve determined that in cases where we detect less than 20% of AI writing in a document, there is a higher incidence of false positives."
https://www.turnitin.com/blog/ai-writing-detection-update-from-turnitins-chief-product-officer
Turn it in has a checkered history and in many ways is the model of surveillance, carceral, intellectual property stealing capitalistic technology.
I am really saddened that we see far less conversation about trusting students, knowing students, and creating student worthy tasks that would largely eliminate the desire or need for these technologies that do virtually nothing to improve student learning.
