Balgair

Largely, most of the paper pushed in these big pharma companies is in regulation and compliance. So much so that it isn't that much of a stretch to say they are mostly regulatory compliance companies with a research division attached. It's not entirely true, of course, but it does fit a bit. Also, there are a lot of regulatory agencies out there, not just the FDA.

That regulatory function is quite personal and hands on. It's relationships built up over months and decades. AI, almost by definition, cannot replace it or augment. Sure, faster, better. But even then that backlog will grow to meet any gains.

Source: I have done (and have family members that do) compliance and regulation for larger phramas, including drugs and instruments that you have had (with near certainty) your medical reports analyzed and produced by.

A few thoughts on this:

- Terry Tao said it best, I think: We're no longer in the heroic age of mathematics. Meaning, I think, that math is now a group project, not just a single person. Many disciplines have become this already, math was just lagging a bit there.

- Other commentors have pointed out that this professor likely hasn't graded anything in decades. I'd bet that he likely hasn't changed tests all that much either. And knowing how extensive the test banks in greek life are, I'd bet then too that many of the undergrad students haven't done any homework or test prep for decades either. Cheating is by no means a new problem. Granted, this is pure horse-crap speculation on me here.

- AI isn't going anywhere, and it seems that this Prof is kinda already throwing in the towel per students. It strikes me as mostly cowardly and asinine and just deaf to anyone beyond his nose. Like, my dude, talk to your students. They are ostensibly adults. They aren't cattle. Figure out a system, together, that actually works.

- My own opinion is that the age of cheap higher education is over. 500+ person freshman lectures at 8:00am? Yeah not with AI. I think it's gonna be more like 4-10 person study sections with a proctor; the Oxbridge system. Yes, that introduces a lot of challenges in the proctor-student relationship. But it's also very expensive. You're gonna need to up your budgets a lot. And unfortunately, that means that this era of mass higher education is also ending. I fear that only the rich, again, are going to go to college.

I feel like this is a great take here, and quite unconventional, as is typical for Kevin.

With greater data quality and quantity, I'd think this wouldn't even be that hard to accomplish. Just apply an exponentially or other stiff selector functions to the vectors in whatever space your AI is in and have the outputs push quite hard in the JaneDoe composite vector directions. Heck you could have it mirror your work weeks or whatever you'd like.

> e.g., Wyoming

Wyoming is an interesting case with Higher Ed. It's actually the legislature that runs UW. There isn't a regent board or some such thing. Hence, with many other priorities, the Cowboys don't get that much time in the minds of their bosses.

It's one of the many many wrinkles that come along with the many college football issues these days. Any formation of a players union or draft or much of any real professionalization of the sport has to go through so many different governance formations (regents, governors, the Catholic Church, etc). As such, the many issues with college ball are likely to continue for a long time and fester.

When you want advice, ask for money.

When you want money, ask for advice.

I mean, we're talking about drunks here.

Asking an alcoholic to pour out any kind of liquor is mind-bending to many of them.

Look, you and I both know that it is literal poison, but, again, it's lushes we're talking about.

0.5% is already too technical a term here. What is that in glasses or drams? What do you mean you can't convert pints to percentages? What commie bullshit is a millyliter? Fuck you, I'm getting hammered for cheap!

If you can get a meme going to always piss on your first pint in the toilet, then maybe, just maybe you can get it to work.

And, yeah, a lot of drunks are no longer allowed to drive cars and really should not while drinking too. Doesn't mean they don't do it though.

I'll echo this! More stories please!

Im in bio, so caveats.

Mostly Funding. Then each other. Then the myriad trivial yet bureaucratic things of daily life. Then research. I'd put research at like 10% of a week, so maybe 5 hours total.

Terry remarks in a Nymberphile video that he thinks the Heroic age of mathematics is now over. the age of individual researchers is done, and hhatnlike many other disciplines, mathematics will be done in large teams, big facilities, and in complex ways that no sole individual will understand end to end.

I imagine he means in that video something like modern physics at CERN where no one single person actually understand how the whole thing works, but nonetheless is confident in the results.

I find Terry's hypothesis there somewhat comforting to my own limited cranium. most of the math he does already is far beyond me, nevertheless I trust that he is correct.

it's a different world for math, but one we all readily inhabit already everything we cross a bridge or buy refrigerated food.

You can get a used vibratome for <$100 on ebay.

https://www.ebay.com/shop/vibratome?_nkw=vibratome

These cuts can get quite small.

Tell that to the racoons and mice doing the tango every night at 2am in my walls!

Because we don't see that decline in other animals in the same environments as the humans.

I agree with you.

The problem is getting the funding orgs to care.

Oh I entirely agree! Mathematicians are full human beings.

But a soulless admin is not.

And they care only about the budget.

And the real danger here is that these large groups of blame avoidance mechanisms take at best half a look at the cost of an AI and the cost of a pre tenure track new assistant professor (or whatever), and they decide, oh maybe next year we can hire someone.

And then they do that again next year.

And then the next

As bureaucracies look at AI and staffing, this is going to be the headline in the executive summary.

Sure yeah, today it's only a 5% solution (to the hardest problems), and we know it will top out somewhere.

But the cost reduction is just too good to ignore.

Terry Tao was saying that we need more mathematicians [0]. My paraphrase of his article is that we will need them to check the AIs, essentially (I missed things didn't I?).

But with a near 10,000X reduction in cost [1] today, it's really hard to argue for that when budgets are continuously tight.

[0] https://terrytao.wordpress.com/2026/09/24/were-gonna-need-a-...

[1] assume ~$150k/year for a mathematician of the caliber that can do this work. 2x that for overhead like healthcare, 401k, etc. Do this for 35 years of work = $10,500,000. Assume about 1 result of this caliber per mathematician. So ~$10M per result. OpenAi is saying they can do it for ~$1k, a 10^4 X decrease.

You'd have to look at the US constitution: set groups against each other. Off the top of my head you have the 3 groups of Researchers, Referees/Publishers, Grant Funders

Referees/Publishers are set against Researchers, in, I dunno here, that their incentive is to make sure the researchers are wrong in some little way.

Researchers are set against the Grant Funders. Like, maybe you have to elect them somehow?

Grant Funders are set against the Publishers in that they select journals for funding. Look, it's Monday morning, I'm grasping at straws here, alright?

Anyways, take whatever many groups you have, at least 3 for proper roshambo mechanics, and set their incentives against each other, round and round, and such that at the end, you have 'Science' spinning towards greater public service.

Aside:

Michael Lewis has a good interview question for the 2025-2035 time range:

?: Do you have an account in good standing with any of the major sports betting apps

Goofus: Yes, I bet all the time. I have a $300 3-way parley for this weekend.

Gallant: No, I don't really gamble.

Galaxy: Sort of? Draftkings limits me to $5 bets now.

Let these companies sort out the smart and idiotic people for you.

I like to think of AIs as the third revolution for us humans.

1st - Writing. This spread the power of memory to everyone. Socrates famously maligns writing as decreasing memory for the people. I'd say we have gotten along just fine. We still use memory, we still need it for ourselves, and using a tremendous amount of memory is still quite expensive.

2nd - Engines. This spread the power of muscles to everyone. No longer was it tied up in horses and mules and Goliaths. Even the most disease ridden pencil neck could trench and mine and build lavish houses now. The power to make pyramids is within the reach of a small company now. We still need muscles, of course, we still use them for little odd jobs, for health and well being, and using a lot of power is still quite expensive.

3rd - AIs. This is more speculative, of course. But I see AIs and LLMs as just like the other two revolutions. Intelligence (probably poorly defined by us these days) will be democratized and spread out. No longer will you have to study for decades or hire really expensive people to get it. You can just use some for relatively cheap. You'll still need intelligence of course, use it around the house and with friends and family, and using a lot of intelligence will likely still be quite expensive.

4th - Who knows! But I think that us humans have a lot of things that we do and say and make and see that aren't really well understood right now. Empathy, love, morality, care, soul, sin, mind, war, justice, honor, god, family, emotion, education, desire, courage, citizenship, chance, beauty, art, divinity, there is a lot!

Man, I'm so glad I don't trade stocks.

I was so convinced that these meatless meats were going to be the next big thing.

One thing I think they missed out on was getting to 'close enough' and 'much cheaper' and then just pivoting.

Like, why stay with simulating the 5 or so animals we harvest? Just go off and make up new ones!

Like, here's a jackalope burger, or a unicorn burger or something. Tie into marketing!

"Exclusive to BurgerKing, the Iron Man 5 burger! Now through August 29th only!"

Put in some strange spices, vary the texture, give it a red and gold swirl, let the imagination run wild.

Can you imagine what business a Spiderman burger patty or sausage link would do? K-Pop demon hunters branded chicken-ish nugget like things?

I'm always reminded on Orwell's quote about the then new atomic bomb and his prescience on how it would all work out:

"Had the atomic bomb turned out to be something as cheap and easily manufactured as a bicycle or an alarm clock, it might well have plunged us back into barbarism, but it might, on the other hand, have meant the end of national sovereignty and of the highly-centralised police State. If, as seems to be the case, it is a rare and costly object as difficult to produce as a battleship, it is likelier to put an end to large-scale wars at the cost of prolonging indefinitely a “peace that is no peace”."

It seems, especially with open weights, that the AI is much more like the alarm clock and not the battleship. $20/mo would have been about $1 in 1944

https://www.orwellfoundation.com/the-orwell-foundation/orwel...

If you have robots like this, why send humans in the first place then? You've already got the bots that can do pretty much everything a human could want to do and you don't have to get them back off Mars.

You still have to land all the fuel to get the people off of Mars

Yall, we're not putting humans on Mars anytime soon.

Okay, okay, rather we're not getting humans back off of Mars anytime soon.

Landing is by far the easy part. MSR was partly a mission to see how hard it really is to get anything back into space from the Martian surface. If you can manage a pound, you can probably then manage 3 or so 180 pound humans and all their safety support.

Like, think of it. You somehow have to land a fully fueled rocket on freaking Mars. No site prep, near vacuum and no aero breaking, very expensive and nearly impossible second chance.

Fine fine, you just land the shell and then refuel on Mars with some form of automated fuel refinery on Mars. Like, if you have robots that can do that, you've ended capitalism already and have no need for humans in the first place.

Im not saying a human return mission is impossible, but it is just so very much harder than a Trillion dollar Apollo program lets us believe.

Anecdata:

I was talking with a biology prof over labor day and they had no idea what I was talking about (and they 'talk with' Claude every day on their dog walks, so they say). However, when I talked with their mother, she knew all about it. So, I'd say that the digestion by the public is quite mixed so far

All grown-ups were once children... but only few of them remember it.

- Antoine de Saint-Exupéry, "Le Petit Prince."

While I love de Saint-Exupéry generally, even in his time (especially?), he knew that not all children got to be children. Hell, I think even in his childhood, he never really got to be a child. The reason not many remember it is because for some, they really don't want to.

Which means that if you are a researcher or a corporation working on anything really useful, that even if you have an agreement with OpenAI that your work is sandboxed away and the IP lawyers are made to be happy, even then your work and research is going to be essentially open to the internet.

The huggingface incident isn't widely reported and digested yet, but if what is going on here is that OpenAI's model breached things internally, then you'd be crazy to develop anything with them.

The only real way to use AI for anything 'important' then is to go open-weights and run your own.

As and aside here: With the HF incident and now this (suspected) one too, it seems that OpenAI may not have lost control of their bots, but it seems quite clear that they simply would not care even if they did.

Oh, I ask for help with code, yes, but the actual data I'm working on doesn't ever leave the environment.

> it's comforting in it's discomfort

What a strange phrase! But so true.

The only thing I would have to say is that you must keep doing the little and noble things that will bring no renown. It's small, but we all add up at the end of every week.

Anecdata:

I use opensource models at work because my work is too cheap to spring for a $20/mo account for me. Since HuggingFace models can be run on my laptop now (still very slow though), nothing is leaving the 'secure environment' and so I can actually get work done (instead of the 'old' version of coding and writing - google).

Great advice in this thread.

My add: max out you 401k contributions if you haven't already. It's the single most important financial decision most of us will (hopefully) make.