▲ 40 ▼ Scientists are using AI to develop superior vaccines against viruses threatening global health (www.cam.ac.uk) submitted 2 days ago by lavenie@lemmy.zip to c/science@lemmy.world 9 comments fedilink hide all child comments
[–] Fallibilist@feddit.uk 9 points 2 days ago (8 children) The people reflexively downvoting everything that has anything to do with AI are something else. permalink fedilink source hideshow 8 child comments replies: [–] neo2478@sh.itjust.works 16 points 2 days ago (3 children) The problem with headlines like this is that it legitimizes slop that comes out of LLMs. It makes it sound like they used something like chatGPT, instead of bespoke machine learning algorithms that were made specifically for vaccine creation. I wish people would stop calling everything AI as if it were only one thing, but I do understand that it comes from a reasonable lack of technical knowledge in the area, and LLM grift companies exploit it. permalink fedilink source parent hideshow 3 child comments replies: [–] Fallibilist@feddit.uk 1 point 1 day ago* (2 children) it legitimizes slop that comes out of LLMs. Does it, though? Seems much more of a Pavlovian response to me. permalink fedilink source parent hideshow 2 child comments replies: [–] micgerne@lemmy.dbzer0.com 7 points 1 day ago It does for a lot of people, yes. I've lost count of the number of encounters I've had with people online or IRL that claim chatGPT is great because it cures cancer or some other such nonsense. permalink fedilink source parent [–] neo2478@sh.itjust.works 4 points 1 day ago I think that for the general population it does. In the sense that they read this story about AI that is being used for medicine, and then they hear criticisms about LLM cconpanies and datacenters, and they might think "well, at least it was being used for medicine" At least that's how I see it with family members and acquaintances I know who fall for the ads for LLMs. permalink fedilink source parent [–] technocrit@lemmy.dbzer0.com 10 points 2 days ago* (last edited 2 days ago) (2 children) And rightly so. The use of the bullshit term "AI" is a red flag for pseudo-science, clickbait, grifting, etc. In this example, it's extremely telling how the article retreats from the clickbait "AI" to the similarly meaningless "AI-guided" and "AI tool", but it never explains where the "intelligence" is. The actual explanation is completely mundane data processing that they would never put in a headline. To design the bird flu vaccine, the team collected the genetic sequence data of all the bird flu viruses logged by surveillance programmes around the world, which take virus samples from wildlife, livestock and humans infected with bird flu. They built an AI tool Actually intelligent humans developed a data processing program to integrate the huge amount of genetic data and use it to design a range of ‘super-antigens’, which contain the antigen features common to this whole group of viruses. These were screened in cells to select the best one for the vaccine, which was then tested in animal trials. Show me a headline that talks about data processing and I'll happily upvote. permalink fedilink source parent hideshow 2 child comments replies: [–] sugarcake@lemmy.zip 2 points 1 day ago Thanks for taking the time to write this. Clickbait is poison. permalink fedilink source parent [–] FatVegan@leminal.space 1 point 1 day ago Don't they know it's called super intelligence now? permalink fedilink source parent [–] username_1@discuss.tchncs.de -1 points 2 days ago But AI is bad and it's bad to write about bad AI! Bad-bad-bad! permalink fedilink source parent
[–] neo2478@sh.itjust.works 16 points 2 days ago (3 children) The problem with headlines like this is that it legitimizes slop that comes out of LLMs. It makes it sound like they used something like chatGPT, instead of bespoke machine learning algorithms that were made specifically for vaccine creation. I wish people would stop calling everything AI as if it were only one thing, but I do understand that it comes from a reasonable lack of technical knowledge in the area, and LLM grift companies exploit it. permalink fedilink source parent hideshow 3 child comments replies: [–] Fallibilist@feddit.uk 1 point 1 day ago* (2 children) it legitimizes slop that comes out of LLMs. Does it, though? Seems much more of a Pavlovian response to me. permalink fedilink source parent hideshow 2 child comments replies: [–] micgerne@lemmy.dbzer0.com 7 points 1 day ago It does for a lot of people, yes. I've lost count of the number of encounters I've had with people online or IRL that claim chatGPT is great because it cures cancer or some other such nonsense. permalink fedilink source parent [–] neo2478@sh.itjust.works 4 points 1 day ago I think that for the general population it does. In the sense that they read this story about AI that is being used for medicine, and then they hear criticisms about LLM cconpanies and datacenters, and they might think "well, at least it was being used for medicine" At least that's how I see it with family members and acquaintances I know who fall for the ads for LLMs. permalink fedilink source parent
[–] Fallibilist@feddit.uk 1 point 1 day ago* (2 children) it legitimizes slop that comes out of LLMs. Does it, though? Seems much more of a Pavlovian response to me. permalink fedilink source parent hideshow 2 child comments replies: [–] micgerne@lemmy.dbzer0.com 7 points 1 day ago It does for a lot of people, yes. I've lost count of the number of encounters I've had with people online or IRL that claim chatGPT is great because it cures cancer or some other such nonsense. permalink fedilink source parent [–] neo2478@sh.itjust.works 4 points 1 day ago I think that for the general population it does. In the sense that they read this story about AI that is being used for medicine, and then they hear criticisms about LLM cconpanies and datacenters, and they might think "well, at least it was being used for medicine" At least that's how I see it with family members and acquaintances I know who fall for the ads for LLMs. permalink fedilink source parent
[–] micgerne@lemmy.dbzer0.com 7 points 1 day ago It does for a lot of people, yes. I've lost count of the number of encounters I've had with people online or IRL that claim chatGPT is great because it cures cancer or some other such nonsense. permalink fedilink source parent
[–] neo2478@sh.itjust.works 4 points 1 day ago I think that for the general population it does. In the sense that they read this story about AI that is being used for medicine, and then they hear criticisms about LLM cconpanies and datacenters, and they might think "well, at least it was being used for medicine" At least that's how I see it with family members and acquaintances I know who fall for the ads for LLMs. permalink fedilink source parent
[–] technocrit@lemmy.dbzer0.com 10 points 2 days ago* (last edited 2 days ago) (2 children) And rightly so. The use of the bullshit term "AI" is a red flag for pseudo-science, clickbait, grifting, etc. In this example, it's extremely telling how the article retreats from the clickbait "AI" to the similarly meaningless "AI-guided" and "AI tool", but it never explains where the "intelligence" is. The actual explanation is completely mundane data processing that they would never put in a headline. To design the bird flu vaccine, the team collected the genetic sequence data of all the bird flu viruses logged by surveillance programmes around the world, which take virus samples from wildlife, livestock and humans infected with bird flu. They built an AI tool Actually intelligent humans developed a data processing program to integrate the huge amount of genetic data and use it to design a range of ‘super-antigens’, which contain the antigen features common to this whole group of viruses. These were screened in cells to select the best one for the vaccine, which was then tested in animal trials. Show me a headline that talks about data processing and I'll happily upvote. permalink fedilink source parent hideshow 2 child comments replies: [–] sugarcake@lemmy.zip 2 points 1 day ago Thanks for taking the time to write this. Clickbait is poison. permalink fedilink source parent [–] FatVegan@leminal.space 1 point 1 day ago Don't they know it's called super intelligence now? permalink fedilink source parent
[–] sugarcake@lemmy.zip 2 points 1 day ago Thanks for taking the time to write this. Clickbait is poison. permalink fedilink source parent
[–] FatVegan@leminal.space 1 point 1 day ago Don't they know it's called super intelligence now? permalink fedilink source parent
[–] username_1@discuss.tchncs.de -1 points 2 days ago But AI is bad and it's bad to write about bad AI! Bad-bad-bad! permalink fedilink source parent
all 9 comments