Stuart Apr 23

500,000 people's data is taken out of UK Biobank and put up for sale on Alibaba: https://www.ukbiobank.ac.uk/news/a-message-to-our-participants-uk-biobank-data-security-update/

UK Biobank demonstrating exactly the reasons I objected to these sorts of "Trusted Research Environments" that seem to give way too much unearned trust to "researchers" assuming that "legal contracts" are good enough to stop bad behaviour. But a contract doesn't stop data being taken. So they are *now* trying to come up with ways to stop large files being extracted from their data centre.

A message to our participants: UK Biobank data security update

We would like to inform you about an incident involving UK Biobank data. We apologise to our participants for the concern this will cause, and we hope to provide reassurance by outlining the serious actions we are taking in response.

UK Biobank
Dan StowellApr 24
Show threadStuart 
Contrast this with Ben Goldacre's OpenSafely which was specifically designed so that researchers never got access to individual's data. Instead they submitted "questions" (with code) that was peer reviewed (in the open) with explicit checks to make sure no PII came out in the answers. Then the code ran inside the data centre and the answers returned. By design it put technical and structural breaks between researchers and PII to avoid these sorts of problem being able to occur.
Apr 23 at 7:31pmWeb
Show threadStuart Apr 23
Here's the Register reporting on the UK Biobank story https://www.theregister.com/2026/04/23/500k_biobank_volunteers_data_listed/
Medical data of 500k Biobank volunteers listed for sale on Alibaba, UK minister reveals

Updated: World's largest biomedical dataset lifted and shifted on Chinese mega marketplace

The Register
Show threadChip ButtyApr 23
@slowe there are so many areas where I'd love this style of sending code to the data