Ulrike Hahn / is on Nostr

Academic @Birkbeck, Univ. of London
Centre for Cognition, Computation, and Modelling

Currently Fellow at Netherlands Institute for Advanced Study working on Digital Democracy with Davide Grossi and Michael Maes

can regularly also be found here: Munich Centre for Mathematical Philosophy, LMU

works on #rationality #argumentation #testimony #socialnetworks #misinformation #computationalsocialscience #Social media for digital democracy

Last Notes

2024-09-20 19:19:58

not a fan of the dramatic pronouncement but I don’t think we are anywhere near having a system in place for defining and pursuing scientific integrity violations that is ready for what is going to happen next…
paper via @jrhawley
https://arxiv.org/abs/2408.06292

2024-09-03 18:57:44

- reply

@npub16w5…dxft and why is that not what ChatGPT4o does, for example during RLHF?

2024-09-03 18:43:23

- reply

@npub16w5…dxft how, though - on your version of 'knowing'?

2024-09-03 18:37:09

- reply

@npub16w5…dxft so your sense of "knowing" is more about 'meta-awareness' (in which case young children and non-human species presumably can't 'know') and not about a particular kind of lawful relationship with outcomes?

2024-09-03 18:37:09

- reply

2024-09-03 17:26:53

- reply

@npub16w5…dxft this is interesting! could you say more about what you mean by 'epistemological component'?

2024-08-28 20:40:54

deceptive behaviour and success on false belief tasks in large language models
https://www.pnas.org/doi/pdf/10.1073/pnas.2317967121

2024-08-15 21:28:19

- reply

@npub15sw…x855 for psych, the problem runs even deeper: often data just *are* the “phenomenon” at the same time as data re-descriptions masquerade as “theory”

2024-08-14 17:03:03

I've written a new blog post on debate about fediverse design.
As always, all comments welcome!
https://write.as/ulrikehahn/building-a-better-fediverse
#fediverse

2024-08-01 22:30:20

new paper regarding the current wider debate about misinformation and disinformation, the threats they pose to democracy, and the possibilities of intervention
https://www.nature.com/articles/s41599-024-03503-6
#misinformation

2024-07-26 01:30:32

DeepMind’s AlphaProof and AlphaGeometry 2, have achieved silver medal performance at the 2024 International Mathematical Olympiad (IMO), solving four out of six problems and scoring 28 out of 42 possible points
https://deepmind.google/discover/blog/ai-solves-imo-problems-at-silver-medal-level/

2024-06-13 06:26:05

in light of the news that the creators of Maven, a new Sam Altman backed social network, thought it was a good idea to ‘ingest’ 1 million posts scraped of mastodon.social without consent, I thought I’d repost something I wrote on GDPR and the Bluesky bridge in February, as much of the discussion about the legal issues directly applies.
https://write.as/ulrikehahn/bridging-to-bluesky-the-open-social-web-consent-and-gdpr

2024-06-08 07:08:28

- reply

@npub1mhz…wteh thank you!

2024-02-25 08:45:49

I wrote something on the #BlueskyBridge and surrounding debate
https://write.as/ulrikehahn/bridging-to-bluesky-the-open-social-web-consent-and-gdpr
I had hope #LawFedi would produce such analyses but I couldn't find anything so there we are...
all comments (and corrections!) welcome!