I do think AI will help, but ironically, it's by stretching an already creaking system to its breaking point and forcing it to reinvent itself. For example, AI is getting to the point where it can produce papers that aren't easily distinguishable from human-authored ones. bsky.app/profile/emol... - ThreadSky

randomwalker.bsky.social • 20 days ago

I do think AI will help, but ironically, it's by stretching an already creaking system to its breaking point and forcing it to reinvent itself. For example, AI is getting to the point where it can produce papers that aren't easily distinguishable from human-authored ones. https://bsky.app/profile/emollick.bsky.social/post/3lhhcf7dmss22

Reposted from Ethan Mollick

Economist @joshgans.bsky.social uses o1-pro to generate a (minor, fun) paper in an hour based on an idea of his, and it gets published in an appropriate peer reviewed journal, with adequate disclosure.

He ends with the same sentiment I am increasingly seeing from fellow academics: what now?

Comments

randomwalker.bsky.social•20 days ago

What happens next? My prediction is that dropping the cost of production to zero will force us to stop attaching value to publication, and look for ways to identify actual intellectual value.

lilianedwards.bsky.social•20 days ago

I dont think so. Metrics always more appealing to non experts than actual outcomes. UK REF supposedly based on expertise already incredibly costly to little outcome. “solution“ will be to get AI to read and summarise the AI slop. Garbage in garbage out. Cant imagine this promotes progress either.

kabirkumar.bsky.social•18 days ago

Trying to make a scoring process for this at ai-plans!

randomwalker.bsky.social•20 days ago

There's also the fact that in many/most fields, there's a limit to AI's impact on production as well. It may be able to do the cognitive parts but the bottleneck may be experiments on humans or some other form of real-world interaction that's hard to automate.

randomwalker.bsky.social•20 days ago

I suspect the AI researchers who are producing this hype are misled by the fact that AI research itself is mostly cognitive/computational, and forget how different most scientific fields are.

2sk22.bsky.social•20 days ago

So AIs will generate academic papers which will then becsummarized by other AIs. Recycled garbage? 😀

davedun.bsky.social•20 days ago

I’m not a scientist, but as a language teacher I am interested in research on lang acquisition, teaching techniques, etc.

This requires people (teachers, students & researchers) performing well-designed repeatable studies, which is notoriously difficult in a classroom setting.

AI won’t help here.

gmac65.bsky.social•20 days ago

Indeed, the deeper question remains "is AI capable of really creative scientific work ?" Personally I'm more inclined to think of it as a stochastic parrot, I don't think LLMs or diffusion models will come close to e. g. cracking the magical recipe for plasma confinement in a tokamak, or anything...

stefanurb.bsky.social•20 days ago

they can help to discern and control plasma instabilities, which is a real help

gmac65.bsky.social•20 days ago

Oh that's good to hear. Do you have a ref to some paper by any chance?

stefanurb.bsky.social•20 days ago

Here is One article about that: https://www.newscientist.com/article/2308243-deepmind-uses-ai-to-control-plasma-inside-tokamak-fusion-reactor/

Comments

Posting Rules

Reply