lizziegibney.bsky.social
Senior reporter at Nature, views my own. Journalist covering physics, AI, policy. Attempting to stop lurking and start posting.
41 posts
2,198 followers
86 following
Prolific Poster
Conversation Starter
comment in response to
post
Agreed! Both projects plan to use (or are already using) humans to verify the errors -- it's a big bottleneck right now
comment in response to
post
See you there! (I'll be in the crowd though 😀)
comment in response to
post
The data in the story was good as of yesterday morning, but today's update from @esa.int now puts the risk at 0.16% -- the impact likelihood is falling very rapidly indeed!
comment in response to
post
Or a question of timing! Yesterday morning ESA's figure was 1.4% & the NASA one was 1.5%. With one more day, it may already have shrunk to 0.16%
comment in response to
post
YR4’s risk was so great that it triggered the first outing of the UN’s rapid-response IAWN group – valuable practice for the future when more such asteroids are likely to be found, and for those times we might not get so lucky.
comment in response to
post
The paywall kicks in after a week. Before then it's free if you register
comment in response to
post
www.nature.com/articles/d41... www.nature.com/articles/d41...
comment in response to
post
Yes the environmental costs of these models is definitely an issue, it's something we've covered a lot @nature.com - a couple of examples here. The cost comes mostly in training, and researchers are figuring out ways to radically improve efficiency, but it's a major problem
comment in response to
post
Correction! As @drpaulwoods.bsky.social has pointed out to me 🙏 this is the first *repeater* that CHIME & its outriggers have pinpointed. The first FRB the telescopes localised was this one, reported in @natureastronomy.bsky.social in Sept www.nature.com/articles/s41... 🔭👏
comment in response to
post
Something we missed in the story? I want to hear from you.
For federal workers or anyone else interested in sharing information, you can reach me securely on Signal at mkozlov.01. We can protect your identity.
comment in response to
post
...and because I always forget 🧪⚛️
comment in response to
post
Academic engagement is currently patchy. Is that because of ethical objections to military/dual-use research? Or because of practical reasons: the restrictions make it too difficult, there is little incentive or scant knowledge about it? I'd love to know your thoughts.
comment in response to
post
Looks like it went under www.graphene-info.com/g6-materials...
comment in response to
post
Good point! So maybe 2025 is the start of the birth of QM 😃
comment in response to
post
I'm all for pushing back deadlines! Why 2027?