YungOnions
@YungOnions@lemmy.world
- Comment on The Ozempic Problem 5 days ago:
P A P A M E A T
- Comment on Exhuma (2024, dir Jang Jae-hyun) 1 week ago:
This was a good film. Takes a while to ‘get going’ but still fun.
- Comment on What movies have you watched this week? 1 week ago:
I watched Robert Egger’s Nosferatu. It was good, not as good as The Witch or The Lighthouse IMO, but still good.
- Comment on Why the 2020 video game Immortals: Fenyx Rising doesn't infringe on the trademark of the 2011 film Immortals? 2 weeks ago:
Immortals was a banger of a film.
- Comment on UK would need forest ‘twice size of London’ to offset new airport expansion. 3 weeks ago:
*…org.uk/…/biggest-ever-tree-planting-year-for-nor…
Some key figures from the 2024 Forestry research summary:
…gov.uk/…/forestry-facts-figures-2024/
-
There are now 3.28 million hectares of woodland in the UK (as of March 2024)
-
Woodland area in the UK now represents 13.5% of total land area
-
20,660 hectares of new woodland were created in the UK in 2023/24 (compared to 12,960 in 22/23)
-
There was a 59% increase in new planting and 28% increase in restocking in the UK in 2023/24
Hopefully that’ll make you feel slightly happier :)
-
- Comment on Why do so many UK electrical sockets have an on/off switch next to them? 3 weeks ago:
Massive if correct
- Comment on If AI spits out stuff it's been trained on 4 weeks ago:
Sexton says criminals are using older versions of AI models and fine-tuning them to create illegal material of children. This involves feeding a model existing abuse images or photos of people’s faces, allowing the AI to create images of specific individuals. “We’re seeing fine-tuned models which create new imagery of existing victims,” Sexton says. Perpetrators are “exchanging hundreds of new images of existing victims” and making requests about individuals, he says. Some threads on dark web forums share sets of faces of victims, the research says, and one thread was called: “Photo Resources for AI and Deepfaking Specific Girls.”
The model hasn’t necessarily been trained with CSAM, rather you can create things called LORAs which help influence the image output of a model so that it’s better at producing very specific content that it may have struggled with before. For example I downloaded some recently that help Stable Diffusion create better images of Battleships from Warhammer 40k. My guess is that criminals are creating their own versions for kiddy porn etc.
- Comment on UK low- to middle-income families far poorer than OECD counterparts – study 5 weeks ago:
- Comment on A Working Man [Trailer] 1 month ago:
Jason Statham is Jason Statham in A Working Man, featuring Jason Statham.