I really did not expect that my industrial email inbox that quickly reaches the same amount of Spam as my academic email. Same same on the other side, just different spam on sale #spam #spamspamspam #eggsbaconandspam

Unpopular opinion: I want to write Fortran code again

Uncomfortable data truth: every dataset has a bias

Is there anywhere a Marvin version of #ChatGTP out there? You know, the depressed and paranoid robot from the Hitchhiker's guide to the Galaxy?
If not, why not?

The UK is talking about Gary Lineker using his freedom of expression and thought instead of the intended policies of its hideous government. That says it all

Inspired thought by a toot from @reid_indeed about the CV highlighting non-success.
I never disliked writing Grant applications in academia, but I hugely disliked writing them knowing about the tiny success chances. This skewed already the joy when writing. And this system set-up ultimately led me to not writing my best applications. They never got my most enthusiastic, best ideas, but overengineered word count optimised, priority area focused technical prose fitted to the funder

I am late to read the Feb ChemistryWorld, but aaaaarrrrrgh .... no, the protein folding problem is not 'effectively solved' by AI also not when we repeat this claim in printed form for the 500th time #BadAItakes #RSC #chemistry

What's not to like with such sceneries on the kid's walk to school

Why I don't want to post on the #Birdsite anymore, exhibit 1001 ...
I am still sad that I lost interactions with some very inspiring people on there (hopefully temporarily) but I really can't support an opinion-maker site owned and run by such a dangerous clown

A recent toot by @L_howes (I am unable to quote-toot) did let me rethink how important it is when using pure ML/AI protocols (particular generative ones) to always remember that they are intrinsically build on correlation only. Causation comes with the builder or user.
Thus generative AI like #ChatBotGPT does not do science, it mimics it. It does not create art, it copies and mimics it. And when we claim a language model makes things up, it really only does what it is trained for