DISCOVERY LOOP: new start-up by Google engineers plan to use AI to automate scientific discovery

[PageofME]

Hah. Maybe But I also wonder if the fact that not everyone does things just for money is one of the differences of opinion here.

I follow AI developments occasionally through The New York Times. Recently, I read Karen Hao’s Empire of AI, which highlighted how the struggle to attract and retain senior talent has turned into an intense salary war.

One of the most interesting facts she presented was that over the past decade, there’s been a sharp drop in PhD-level specialists remaining in academia—mainly because tech companies have essentially bought them all up.

 

[hotblack]

I follow AI developments occasionally through The New York Times. Recently, I read Karen Hao’s Empire of AI, which highlighted how the struggle to attract and retain senior talent has turned into an intense salary war.

Yes I’m familiar with her work and perspective and there is undoubtedky a lot of daft money and poaching of staff going on in places. But there are also people leaving companies to set up start ups or other ventures because they believe in something else, be it a vision for how the technology should be developed or used. Plenty of people here as in other fields haven’t used the biggest salary they can get as their sole motivation.

 

[hotblack]

One of the most interesting facts she presented was that over the past decade, there’s been a sharp drop in PhD-level specialists remaining in academia—mainly because tech companies have essentially bought them all up.

That happens in lots of fields as they move from research to having commercial applications. And that has happened here, there are more opportunities in commercial operations than there used to be. It’s a trend. Things will change over time.

As an example, many maths graduates have been bought up by finance companies for years. But not all maths graduates go for the biggest cheque. I know plenty of v smart people who could have earned more doing x but chose to do y instead.

 

[EndME]

As an example, many maths graduates have been bought up by finance companies for years. But not all maths graduates go for the biggest cheque. I know plenty of v smart people who could have earned more doing x but chose to do y instead.

And many of those that leave, don't actually leave because of the larger cheques but for example because of other hurdles in academia (no permanent positions being available in a given city etc.) or because they've accomplished everything they can.

I do however suspect that things will change quite drastically in the upcoming years as a majority of scientific professions are completely taken over by AI. Going back to the article, I don't see why humans would still be in labs much longer. Things can change quite quickly.

 

[Yann04]

Yes I’m familiar with her work and perspective and there is undoubtedky a lot of daft money and poaching of staff going on in places. But there are also people leaving companies to set up start ups or other ventures because they believe in something else, be it a vision for how the technology should be developed or used. Plenty of people here as in other fields haven’t used the biggest salary they can get as their sole motivation.

Also if you are genuinely one of the top ML/AI/LLM experts and have been working somewhere like Google for a couple years, you will probably be so obscenely rich in equity that you’ve basically guaranteed an upper class lifestyle for the rest of your life, so for some money starts mattering less

 

[PageofME]

This seems to be an interesting podcast around AI and medicine in general practice.

Overall, I think the internet has already been a blessing for patients to connect and question theories and therapies that don't work—like in ME/CFS. It's also leading to an empowerment of patients in psychiatry and psychotherapy, provided you are actually trained or skilled enough to find out what describes your medical problem best and what the most beneficial thing to do is.

When it comes to how doctors respond to these developments in media and information technology, there seem to be different approaches. Some doctors value patient engagement because they genuinely want to help, and when they see patients benefiting, they're happy too. Others are simply getting overwhelmed by how fast everything is developing—like many non-doctors, too.