retropz.bsky.social
Research Integrity and Training Adviser at the Cancer Research UK Manchester Institute
Interested in improving #ResearchIntegrity & #ResearchCulture with training, #TeamScience & promoting #OpenResearch
https://orcid.org/0000-0002-3353-7002
1,065 posts
2,138 followers
3,425 following
Regular Contributor
Active Commenter
comment in response to
post
📌
comment in response to
post
"It has been said that history repeats itself. This is perhaps not quite correct; it merely rhymes."
Theodor Reik (often attributed to Mark Twain: quoteinvestigator.com/2014/01/12/h...)
comment in response to
post
Another parallel might be GMOs, where national/international regulation was put in place to safeguard against dangers (and potential misuse by companies, e.g. terminator seeds needing re-purchasing), but which probably hindered the commercial uptake of the tools by decades (worthwhile or not?)
comment in response to
post
Great point! Two reasons I'd like to explore the PFAS parallel some more are 1) it seems companies knew at least some of the dangers to human health but nonetheless kept producing and marketing the products and 2) the pollution of the environment is long-lasting and very hard to reverse
comment in response to
post
One technology that GenAI is reminding me of is PFAS - en.wikipedia.org/wiki/PFAS - the 'forever chemicals' which provided amazing benefits (Teflon, waterproofing, new materials etc) but which are potent pollutants & toxins
The science around these issues lagged way behind the tech deployment
comment in response to
post
I reckon I think about this single image as much as some people claim to think about the Roman Empire
comment in response to
post
It's not my field or method of research, so I don't have comparisons on how rigorous this is.
But getting feedback from an author of each paper on how good the #AI podcast is feels like only one component - I'd like to see what sci comm professionals, other researchers and the public themselves say
comment in response to
post
Maybe they're like Mark Watney in 'The Martian' and Disco is all they have available on their playlist?
comment in response to
post
I get that people want to experiment with new technologies & have no issues with running tests to see if those work
But we know enough about the current state of this tech to know that turning a paper into a podcast for patients is not the same as medical communications experts creating podcasts
comment in response to
post
Also (and appreciate this is not your main point, which I agree with) calling an AI intended to streamline the civil service and offer unbiased advice to ministers 'Humphrey' is either a complete misreading of 'Yes, Minister' or some high-level trolling of the government
comment in response to
post
[Long awaited John Le Carre sequel title revealed]
comment in response to
post
Shame it was before the days when you could pop into Café Nero and use their WiFi
comment in response to
post
comment in response to
post
🎯 I scored 5/10 on #DiscoverOrMDPI
pagoba.shinyapps.io/publi_guess/, #ResearchIntegrity #ScientificPublishing #AcademicSky
comment in response to
post
Bookmarking this as a very helpful example of how to use the REAPPRAISED checklist 📌
comment in response to
post
📌
comment in response to
post
📌
comment in response to
post
Bullseye 🎯
comment in response to
post
It's also interesting that once the AI starts producing content that is accurate - i.e. the response 'no I didn't create this' - they now say 'its telling the truth and apologising' - as if *this* content is in anyway different than what came before - i.e. probabilistic responses to inputs
comment in response to
post
📌
comment in response to
post
comment in response to
post
comment in response to
post
The wonders of syndicated television and a large pool of writers!