Ben Williamson / is on Nostr

Researching and writing about education, digital tech, data and policy at the University of Edinburgh | Co-editor of Learning, Media & Technology | Code Acts in Education https://codeactsineducation.wordpress.com/

Last Notes

2024-08-23 21:30:19

Hey, if you are or know teachers up against pressure to use AI, and you or they are feeling uneasy about it, here are two AIED-critical resources from last term I hope are helpful: (1) a report on the harms of AI in schools https://nepc.colorado.edu/publication/ai

2024-08-16 23:13:01

Has your paper been used to train an AI model? Almost certainly". Some on here say we brought it on ourselves (while boosting their AI ideals). They don't get how infrastructural academic publishing is to academic careers, and thus how exploitative this feels. https://www.nature.com/articles/d41586-024-02599-9

2024-03-09 00:33:15

- reply

@npub1xr8…3q0g all I can say it's a disaster - all our report asks for is a pause and reset before more damage is done. But I doubt our own recommendations will achieve anything - at least we tried, which few others are doing.

2024-03-08 23:38:03

- reply

@npub1xr8…3q0g seems encouraging. Here it's just we'll solve austerity with chatbot teachers

2024-03-08 23:20:15

Is AI inevitable, necessary or good for schools? Short thread on our new report about AI in schools written for schools and leaders
tl;dr question industry and investor insistence that unregulated, untested AI is inevitable, necessary or good for education https://nepc.colorado.edu/publication/ai

2024-02-10 23:53:26

Could Cory Doctorow's neat "enshittification" concept apply to edtech? I can think of a good example. A little thread...

2024-02-09 21:56:27

- reply

@npub1ahr…m294 thanks, yes, if we had resources we'd do more but the objective here was to understand the knowledge making infrastructure. Should be more to come on the knowledge claims.

2024-02-09 20:50:34

How has it become possible to claim educational outcomes can be predicted from genetic data? Our new paper from our Leverhulme Trust-funded BioEduDataSci project on "Infrastructuring educational genomics" tries to answer https://link.springer.com/article/10.1007/s42438-023-00451-3

2023-11-02 19:19:00

Super special issue intro on "Digital Academia" for anyone wanting to put current speculations about "digital transformations" - like AI and big data - in higher education into context https://sciendo.com/article/10.2478/sjs-2023-0022

2023-10-07 23:38:16

Pretty much peak AI is tech companies surveying tech executives on what skills everyone needs educating in, and finding it's "AI" skills, then launching free online courses to habituate millions - especially "underrepresented" groups - to working with AI https://fortune.com/education/articles/ibm-commitment-two-million-workers-in-ai-by-2026/

2023-07-01 18:44:19

- reply

@npub1ahr…m294 @npub1mkr…htal maybe but the piece is about "AI" and its specific problems - which are often reproductive of existing problems. These seem like reasonable things for all us educators to consider seriously, based on what we are coming to understand about "AI", rather than simply dismiss as nothing to worry about, no?