Alongside warnings about its ability to destroy humanity, artificial intelligence is also hailed for its potential to revolutionise health treatment. Just how far can it take us?
AI researcher here: cancer research – and all other research as well – has had access to AI for decades. This includes LLM models. It is doubtful that commercial models built on copyright infringement and Reddit content will play a key role in any cancer research breakthroughs.
Big AI companies keep pushing this narrative to silence critical voices.
If cure for cancer is something you wish to see in the future, my advice is to support universities in any way you can (funding being an obvious method). It is the job of universities to advance science in a manner that benefits everyone. Medical companies’ job is to serve shareholders’ money only. Its sad, but its true because its written in the law.
Taking this and your other comment into account, your point of view is very much justified. Two things:
The history of AI research has had many “AI winters”, where belief in the usefulness of AI tech has diminished, reducing the amount of funding (and increasing the amount of ridicule) for AI research. “AI effect” is a related term here – it means that the goalpost keeps moving further as we learn that what is sold as AI was instead just a clever algorithm. I’ve been wondering why this term isn’t making rounds again.
“AI” is a misleading term that maybe should be abandoned. But instead of abandoning it, Big AI has stuffed it with steroids. There was a moment of dignity in AI research circles when we didn’t envision a Star Trek future, but instead agreed that AI was a class of algorithms with no real intelligence included. Now it seems that you can be an “AI engineer” after two weeks of product training and qualified to tame the Skynet that is lifting it’s malicious, cunning and indifferent head.
When I refer to AI, I mean what we had in the moment of dignity right before OpenAI and Anthropic burst into the scene. AI algorithms lacked intelligence back then, just as they do today. I wish people had more respect for real human intelligence.
This — but, I’m sure at some point, AI could develop a cure for various cancers. If it’s a straight-up race between humans and AI, the AI will win. It doesn’t need to rest and it can do more, faster. I do not see this as being up for debate. If the human finds one cure first, it will be luck or it will be some dumb thing that the AI overlooked. Something like that.
But yes, if the companies paying for AI get the cure, they will not share it with the world, they will sell it at a very high price.
Most rich people aren’t stupid. Some were given their money, like the US president, but a lot of them are also educated. They realise the Earth’s population is not sustainable. They will happily let poor people die to create a better world for themselves. The cure will go to the rich and powerful, and they’ll let the working poor fight over the scraps.
Altruistic researchers should be the ones to discover the cures, but they probably won’t be. And even if they discover one or two, it doesn’t change much. Cancer isn’t one disease, it’s a family of diseases and a cure for lung cancer won’t magically cure brain cancer or breast cancer, unless we cure the root of all cancers — and this is where my education and experience fall short of someone who actually knows what they’re talking about here — which I believe is its uncontrollable reproduction. So I think if we could stop cancer cells from multiplying, that would go a long way toward treatment, but I would have to defer to the actual scientists and doctors here.
We don’t have AI though. We have LLMs which are glorified auto correct machines that generate globs of text based on which words are most likely to appear near which other words.
These “AI” tools are not the future of research. Sure, sometimes they get lucky and solve a complex problem, but most of the time they spew out useless junk that no one needs.
When the AI bubble pops, all of these LLMs will die off. A few will remain here and there, but they will be a shadow of what they are now.
Machine learning is what will, and already has, change the medical research field. Specialized algorithms to process mountains of data that would take humans ages to do. Probably the only good thing to come out of this AI bubble is all the research and advancements poured into machine learning.
AI researcher here: cancer research – and all other research as well – has had access to AI for decades. This includes LLM models. It is doubtful that commercial models built on copyright infringement and Reddit content will play a key role in any cancer research breakthroughs.
Big AI companies keep pushing this narrative to silence critical voices.
If cure for cancer is something you wish to see in the future, my advice is to support universities in any way you can (funding being an obvious method). It is the job of universities to advance science in a manner that benefits everyone. Medical companies’ job is to serve shareholders’ money only. Its sad, but its true because its written in the law.
*psychic damage*
Allergic to the term as I am or did I say something funny?
Models models
Thanks. I see my mistake now.
ATM machine. CPR resuscitation. GPS system. PCB board. IT technology. AED defibrillator. PIN number.
AT Machine. CP resuscitation. GP System. PC board. I Technology. AE Defibrillator. PI Number.
Better :) ?
Ah, I see. Sorry for the psychic damage.
It’s not your fault. You can hide behind the “colloquial” language shield :P
A good showcase of the AI Effect.
Taking this and your other comment into account, your point of view is very much justified. Two things:
When I refer to AI, I mean what we had in the moment of dignity right before OpenAI and Anthropic burst into the scene. AI algorithms lacked intelligence back then, just as they do today. I wish people had more respect for real human intelligence.
This — but, I’m sure at some point, AI could develop a cure for various cancers. If it’s a straight-up race between humans and AI, the AI will win. It doesn’t need to rest and it can do more, faster. I do not see this as being up for debate. If the human finds one cure first, it will be luck or it will be some dumb thing that the AI overlooked. Something like that.
But yes, if the companies paying for AI get the cure, they will not share it with the world, they will sell it at a very high price.
Most rich people aren’t stupid. Some were given their money, like the US president, but a lot of them are also educated. They realise the Earth’s population is not sustainable. They will happily let poor people die to create a better world for themselves. The cure will go to the rich and powerful, and they’ll let the working poor fight over the scraps.
Altruistic researchers should be the ones to discover the cures, but they probably won’t be. And even if they discover one or two, it doesn’t change much. Cancer isn’t one disease, it’s a family of diseases and a cure for lung cancer won’t magically cure brain cancer or breast cancer, unless we cure the root of all cancers — and this is where my education and experience fall short of someone who actually knows what they’re talking about here — which I believe is its uncontrollable reproduction. So I think if we could stop cancer cells from multiplying, that would go a long way toward treatment, but I would have to defer to the actual scientists and doctors here.
We don’t have AI though. We have LLMs which are glorified auto correct machines that generate globs of text based on which words are most likely to appear near which other words.
Probabilistic scrabble isn’t going to cure shit.
To be more exact, if probabilistic approach could have found the cure for something, it has found it by now.
Breakthroughs typically require a new way of thinking about the problem, not throwing more brute force at it.
LLM = AI
AI ≠ LLM
What Big tech is currently pushing are LLMs, not the type of “AI” that could potentially find a cure. Hence the purpose behind my specificity.
https://www.math.net/symmetric-property-of-equality
This isn’t a mathematical equation.
All squares are rectangles, but not all rectangles are squares. AI is the rectangle and LLMs are the squares.
All spaghetti is pasta.
These “AI” tools are not the future of research. Sure, sometimes they get lucky and solve a complex problem, but most of the time they spew out useless junk that no one needs.
When the AI bubble pops, all of these LLMs will die off. A few will remain here and there, but they will be a shadow of what they are now.
Machine learning is what will, and already has, change the medical research field. Specialized algorithms to process mountains of data that would take humans ages to do. Probably the only good thing to come out of this AI bubble is all the research and advancements poured into machine learning.