My guess is the aversion is real, and maybe somewhat cynically I see the rejection of phone calls and emails as motivated mostly by rebellion/fashion cycles rather than anything else. Phone calls and emails just aren't cool anymore. Your parents used them, yuck! And all the boring, very uncool workplaces and companies require them. Very similar to how gen-z dislikes Facebook in favor of Instagram. Facebook is where the boomers are.

The book in which Alex Wellerstein really makes the case was also released yesterday, buy it here!

There is also recent debate about whether Truman was even well informed about the fact that Hiroshima was a city rather than a "purely military target", eg see the book The Most Awful Responsibility, well reviewed by many including Richard Rhodes, as well as the excellent interview with the author by Dan Carlin.

There is a dis-analogy, in the former case you have a single goal, get good at chess. In the latter case there are many goals we want AIs to do, ranging from coding to running scientific experiments to curing diseases and even making art. Obviously if you want a generalist you will want to teach general skills.

Secondly, a big reason labs are focusing on ML research is to get on the recursive-self-improvement super-exponential curve.

Your analogy addresses neither of these points, and I do think that these points are the primary reasons why people are trying to get AIs to do well at ML research. Therefore I think your analogy is bad and you should not make inferences or plans using this logic.

I feel like there was a time just after ChatGPT became big where there were so many different new & competing frameworks for what exactly was going on, shard theory and simulators in particular but also active inference the MIRI views and all the people rolling their own ontologies with which to understand LLMs, and notably trying to make those ontologies explicit and comparing them against each other.

Maybe I'm just around very different people now or doing very different work than I was (I am on both counts), but those conversations aren't really happening anymore. I don't know whether its for good or ill all things considered, but I do get nostalgic for them sometimes.

Interestingly, my tweets were (for whatever reason) much better than usual during this period.

Presumably because you were drunk and angry

The progress I listed doesn't seem like it's going slower due to medical regulation.

I mean the basic research aspect sure (except for stem cells), but applications of each of the progress areas you listed basically involve either clinical applications or selling GMOs. Both of which have very bad regulatory bottlenecks, especially from a world-wide perspective.

There has been, as you mention, enormous progress in bio-tech and our broader understanding of biology in the past 50 years, but comparatively little application of that knowledge. This is not what you would expect if the science is "deep" but applications easy. How exactly does the progress you listed support this conclusion?

My guess is that the big difference in the speed of biotech compared to early-20th-century-advancements is the relative conservatism of the medical field, and the money & time-consuming certifications you need to get before releasing anything to market. This, in my view, is much less a function of the science, and much more a function of the sociology around the science.

so yup np-complete. are they halting oracles?

You may be interested in Scott Aaronson et al's paper on the subject of computability theory of closed timelike curves

We ask, and answer, the question of what's computable by Turing machines equipped with time travel into the past: that is, closed timelike curves or CTCs (with no bound on their size). We focus on a model for CTCs due to Deutsch, which imposes a probabilistic consistency condition to avoid grandfather paradoxes. Our main result is that computers with CTCs can solve exactly the problems that are Turing-reducible to the halting problem, and that this is true whether we consider classical or quantum computers. Previous work, by Aaronson and Watrous, studied CTC computers with a polynomial size restriction, and showed that they solve exactly the problems in PSPACE, again in both the classical and quantum cases.

Compared to the complexity setting, the main novelty of the computability setting is that not all CTCs have fixed-points, even probabilistically. Despite this, we show that the CTCs that do have fixed-points suffice to solve the halting problem, by considering fixed-point distributions involving infinite geometric series. The tricky part is to show that even quantum computers with CTCs can be simulated using a Halt oracle. For that, we need the Riesz representation theorem from functional analysis, among other tools.

We also study an alternative model of CTCs, due to Lloyd et al., which uses postselection to "simulate" a consistency condition, and which yields BPP^path in the classical case or PP in the quantum case when subject to a polynomial size restriction. With no size limit, we show that postselected CTCs yield only the computable languages if we impose a certain finiteness condition, or all languages nonadaptively reducible to the halting problem if we don't.

Re: Politicians: Andrew Yang isn't a major politician I guess, but his main schtick was "AI is coming" basically right?

Not really, from my memory and checking wikipedia, his campaign was mainly focused on advocating for UBI, and used whatever arguments it could to defend that policy position, including but certainly not limited to an argument that automation was coming, but mainly for menial tasks like truck driving.