AI-designed universal vaccine clears first human trial, targets future coronavirus threats with needle-free delivery.
https://medicalxpress.com/news/2026-06-ai-universal-vaccine-human-trial.html
https://medicalxpress.com/news/2026-06-ai-universal-vaccine-human-trial.html
31 Comments
RobotToaster@mander.xyz · 40 pts · 59d
Universal coronavirus vaccine, a vaccine against everything would induce all sorts of autoimmune disorders.
Natanael@slrpnk.net · 6 pts · 59d
There's another universal kind-of-vaccine in development, which doesn't tell your body what to recognize but instead tells your body how to react (primes the immune system to respond faster to respiratory infections with localized antibodies, etc)
tristynalxander@mander.xyz · 23 pts · 58d
Evotech@lemmy.world · 5 pts · 58d
It's not gen ai. It's ai in the more traditional sense
darkangelazuarl@lemmy.world · 2 pts · 57d
Yeah, Protein design is absurdly complicated and time consuming. Collaborative computing programs like folding at home have helped a lot but its still like trying to find a grain of sand in a desert. Having a well tuned AI help narrow down what needs additional attention via other computer models or even moving past computer models is a reasonable use of a well tuned specific AI. Not these general purpose LLMs that just tell you what you want to hear.
PotatoPie@lemmy.zip · -4 pts · 58d
I'm anti AI but i have to give it to the technology, for once, it's used for something good and for a task no one wanted to manually do, with incredible results
Until the sector is regulated i'll just say this never happened
gedfromgont@piefed.ca · 4 pts · 58d
Technically this probably isn't AI but rather ML (machine learning) which is something that made great strides in science already in the past decade or so. E.g. alphafold, a program to predict folding of protein sequences predates all LLM crap and has become a cornerstone of anything protein expression.
tristynalxander@mander.xyz · 2 pts · 57d
gedfromgont@piefed.ca · 3 pts · 57d
So when I talk about LLM, I mean in first line ChatGPT since it was the big event bringing LLMs to the public.
Alphafolds first version came out in 2018, ChatGPT was only released in 2022. I don't know where AI image generation comes into this, but that wasn't my point anyway.
tristynalxander@mander.xyz · 1 pts · 57d
wrinkle2409@lemmy.cafe · 2 pts · 57d
Machine learning is categorically part of AI. It is just that people started slopply using the term to mean exclusively LLMs and generative models. The problem has never been AI but rather who's using it
aeronmelon@lemmy.world · 19 pts · 59d
Oh, right. The vaccine. The vaccine for COVID. The vaccine chosen specifically to kill COVID. COVID’s vaccine. That vaccine?
WolvenSpectre@lemmy.ca · 9 pts · 58d
No, because Vaccines don't kill Covid. Vaccines teach your immune system to respond better and on the whole have better outcomes. If vaccines killed Covid, then they would give a shot to everyone with Long Covid and cure them.
passenger@sopuli.xyz · 1 pts · 57d
Afaik long covid is caused not by covid being still in the person but rather brain / lung / etc. damage it has done.
Anyways vaccines don't kill the virus so you are right.
Viruses aren't alive anyways...
Gladaed@feddit.org · 17 pts · 59d
Note that AI has been a cornerstone in high dimensional optimization for a while.
bl4ckp1xx13@lemmy.dbzer0.com · 15 pts · 59d
Useful case of AI. 👏
whats_a_lemmy@midwest.social · 17 pts · 58d
The original press release stated this was done using machine learning, which has been a mainstay in biomedical science for a while. It's what "AI" used to refer to before the current generation of bullshit generators.
bl4ckp1xx13@lemmy.dbzer0.com · 0 pts · 58d
Yes, but from what I understand traditional machine learning is more about introspective pattern recognition and not necessarily inference-based synthesis?
a_non_monotonic_function@lemmy.world · 1 pts · 58d
Those are certainly word.
Zachariah@lemmy.world · 11 pts · 58d
it’s the stringent verification of results that makes it useful
Gladaed@feddit.org · 5 pts · 58d
That's true for all research in medicine. The suggestion to not check for errors is mental and would only be suggested by non technical people.
Zachariah@lemmy.world · 1 pts · 58d
unfortunately, this how most companies are run
Gladaed@feddit.org · 1 pts · 58d
I am sorry, but did you read that I was talking about medicine first and foremost?
Also using evidence based approaches can be beneficial for longevity of any company and if you truly believe otherwise, then that's your perogative.
Zachariah@lemmy.world · 1 pts · 58d
Yes, that’s why I pointed out why it works in this field.
The way people run companies in others fields is mental, and they want to use AI without verification. They shouldn’t, but they do (hence the “unfortunately”). And even when they say they’re evidence based, it’s usually cherry picked to justify what they were going to do anyway.
Despair@lemmy.world · 5 pts · 58d
It's a jet injector.
noxypaws@pawb.social · 1 pts · 58d
I'm curious about this too. Surely they're using some modern advancements in the tech that have addressed that issue, SURELY.
evenglow@lemmy.world · 10 pts · 58d
DarrinBrunner@lemmy.world · 6 pts · 58d
I'm all for machine learning used this way. As long as the trials are as stringent as with any other method.
SaveTheTuaHawk@lemmy.ca · 6 pts · 58d
Safe with no side effects does not mean that it works. A glass of water is safe with no side effects.
Omgpwnies@lemmy.world · 1 pts · 57d
A glass of water can kill you if ingested wrong, say no to DHMO!
CultLeader4Hire@lemmy.world · 3 pts · 58d
Will anti vaxxers find AI formulated vaccines more or less abhorrent?
Snapz@lemmy.world · 4 pts · 58d
They don't know yet, they haven't been told what to feel.
Bristlecone@lemmy.world · 1 pts · 58d
Great question. Most of them probably more so I'd imagine still
Evotech@lemmy.world · 1 pts · 58d
Imagine making that avd getting reduced to "ai designed" lol