Kate Devlin
@drkatedevlin.bsky.social
Academic; writer. Professor of AI & Society, Chair-Director @kings-dfi.bsky.social, King’s College London. Come for the sex robots; stay for the eye-rolling at AI nonsense on a daily basis. #academicsky . Norn Irish in Norwich.
Unconferences are my favourite. The beautiful serendipity of it all.
November 11, 2025 at 10:46 PM
Unconferences are my favourite. The beautiful serendipity of it all.
Congratulations! ❤️
November 10, 2025 at 8:21 AM
Congratulations! ❤️
The absolute truth in this.
November 9, 2025 at 5:44 PM
The absolute truth in this.
It's like a solipsistic parody.
November 9, 2025 at 9:01 AM
It's like a solipsistic parody.
Obviously this makes me sound old and out of touch and grumpy. Fine.
November 9, 2025 at 9:00 AM
Obviously this makes me sound old and out of touch and grumpy. Fine.
My da is never happier than when he is complaining about something, particularly if he is complaining about something that has little to no effect on his day-to-day life.
(Also, a WEEK? At a time? You’re a saint.)
(Also, a WEEK? At a time? You’re a saint.)
November 6, 2025 at 8:07 PM
My da is never happier than when he is complaining about something, particularly if he is complaining about something that has little to no effect on his day-to-day life.
(Also, a WEEK? At a time? You’re a saint.)
(Also, a WEEK? At a time? You’re a saint.)
My most controversial tweet ever was me saying we should license metal detectors. The absolute fury it unleashed! I think this is a great idea.
(I am also in favour of licensing poets.)
(I am also in favour of licensing poets.)
November 6, 2025 at 9:39 AM
My most controversial tweet ever was me saying we should license metal detectors. The absolute fury it unleashed! I think this is a great idea.
(I am also in favour of licensing poets.)
(I am also in favour of licensing poets.)
AND THEN when I tell people “I rarely use LLMs, maybe a couple of times a month to reduce word counts or rephrase something for boring copy” they say “well, how can you understand it if you don’t use it?” (Short answer: a PhD in computer science.)
November 4, 2025 at 8:18 AM
AND THEN when I tell people “I rarely use LLMs, maybe a couple of times a month to reduce word counts or rephrase something for boring copy” they say “well, how can you understand it if you don’t use it?” (Short answer: a PhD in computer science.)
I am utterly fed up with “well, it’s quite good for initial research” when it’s palpably not. It’s quite good at sounding good about things you don’t know enough about to handle critically. I fucking despair. And I agree with you about escape rooms.
November 4, 2025 at 8:16 AM
I am utterly fed up with “well, it’s quite good for initial research” when it’s palpably not. It’s quite good at sounding good about things you don’t know enough about to handle critically. I fucking despair. And I agree with you about escape rooms.
I had this from someone who told me “ChatGPT said these are the updates you need to make to your published research.” Not a single one was valid. I had to explain both the nuances of my research subject and how large language models actually work.
November 4, 2025 at 8:13 AM
I had this from someone who told me “ChatGPT said these are the updates you need to make to your published research.” Not a single one was valid. I had to explain both the nuances of my research subject and how large language models actually work.