Tag Archives: scientific progress

April 2026 in Review

Leave a reply

In fast-moving current events as I write (Saturday, May 2), active so-called “kinetic” warfare seems to have subsided in and around Iran. Let’s hope the trend continues in this hopeful direction. Human rights violations elsewhere and global economic impacts persist.

Most frightening and/or depressing story: We have heard horror stories about U.S. government debt over the decades, many not grounded strictly in evidence. But this time really seems to be different, where the absolute size of the debt at the moment means higher than normal interest rate payments as a fraction of the economy and tax revenue. At the risk of stating the obvious, this means the government has less money for things other than interest payments. Meanwhile the trends are increasing debt level, increasing interest rates, and potentially lackluster economic and tax revenue growth, all pointing toward a runaway train. Hoping for a pickup in economic growth seems to be the main strategy being pursued to counteract this feedback loop.

Most hopeful story: AI science seems to have been a theme of mine in April. We can constraint an AI scientist to actually respect the laws of physics, potentially accelerating scientific and technological progress. AI should also be good at synthesizing past research to form a basis for future progress, and organizing data in an accessible way so that others (human and/or AI) can confirm findings or make new discoveries from that same data. I know some very nice people who work in today’s academic publishing industry, but this may not be an area of rapid future growth. The future of engineering and scientific modeling will probably consist of giving an AI a very detailed specification for what you want it to accomplish, then reviewing/validating the result when it comes back.

Most interesting story, that was not particularly frightening or hopeful, or perhaps was a mixture of both: Augmented (aka mixed) reality glasses are getting pretty common in China, and slowly catching on elsewhere. Early adopters include cheating students, of course.

Academic publishing beyond the current citation regime

Leave a reply

This blog post from someone named David Oks pines for a time before citations became a central pillar in the effort to advance scientific publishing. Thank you Mr. Oks for sharing your views, which got me thinking. I am not sure I agree 100%, because an important part of civilization and science is our ability to document the state of knowledge so that each successive generation can build on it. But the point is taken that the extreme focus on gaming this one metric (number of citations, citation scores of individual academics and journals) has become an end in itself, rather than a means for advancing civilization and science.

I do think AI can be very good at improving the state of the “literature review”. Every scientific article starts with a summary of literature on the topic, which the authors then typically build on (although, some articles are just literature review). A human author can spend years sifting through a vast amount of literature relevant to a topic, particularly a novel or interdisciplinary topic, trying to find those few needles in a haystack that are really the most, say, 100 highly relevant papers, and synthesize them into a foundation that can be built on it. I have done this, and it is actually a very fun thing to do (for my personality type, I suppose, not for everyone, but there are many like me…) Through this process, you gradually build and refine your own unique mental model on a subject which then can become the foundation for your personal unique contribution to human progress. However, even spending years, you can’t come close to looking at everything possibly relevant, and you can miss some of those needles. An AI should be able to look at literally the entire haystack, find and synthesize the needles, in a matter of minutes or at most hours, compared to months or years for a person. The only sad thing here is that the best mental models in the brains of the smartest humans might not be built in the same way. Overall though, I suspect the rate of progress can be increased significantly.

What is an alternative to the citation regime according to Mr. Oks?

So I suspect that we’ll have to fundamentally rethink the institutions of scientific life for the age of strong AI. Perhaps, as AI makes it possible to do much more science much more quickly, the culture of science will become more like the culture of engineering—faster, more collaborative, less interested in priority claims. In such a world, the most efficient unit of scientific contribution might be a living document, perhaps even just a GitHub repo: something with data, code, analysis, and a thin narrative layer that AI scientists could read, regenerate, or update as needed. And citations, in this world, could ultimately become obsolete. Journal articles would survive, though perhaps they’d become something closer to definitive pronouncements on major breakthroughs or on the state of knowledge in a given domain—a bit like what scientific books were before the rise of journals. In a world where science is much more productive than it is today, legitimacy will be the scarce factor in the production of useful scientific knowledge.

I am not sure I have experienced the culture of engineering described here in my own engineering career, but this may be referring to something more like “product development” rather than the public infrastructure and environmental planning type work I do. Anyway, my idea of AI-backed knowledge synthesis plus this idea of open data and code may be on to something.

instructions for the AI scientist

Leave a reply

Here’s an example of how AI science could work. If you ask machine learning to review a data set from a system and predict the future behavior of the system, it can often do a good job, but the mathematical approach it is using to do that is mostly opaque. It is also divorced from any sense of the system structure, its interaction with the physical universe, and the physical laws governing it. But you can ask the computer to constrain itself within those known physical laws, and then it may be able to provide you insights on the structure and physical processes inside the system. So at that point, you and/or the computer should be able to form scientific hypotheses and test them against the data. This example from Water Resources Research is about soil moisture.

Coupling Intuitive Physics Into Deep Learning for Soil Moisture Flow Processes Learning

Soil water flow processes in the unsaturated zone support ecosystems and regulate water, energy, and biogeochemical cycles. Recently, deep learning (DL) approaches have significantly advanced soil moisture (SM) prediction tasks yet still challenging to interpret. It’s difficult to peer into the internal reasoning procedures of algorithms, let alone associate them with specific physical processes. Thus, DL alone is unlikely to satisfy soil hydrological modeling needs and cannot advance process understanding. Here, we present DPL-S (deep process learning for SM dynamics) approach, which couples intuitive physics into deep learning architecture as structural guidance to facilitate comprehensive surrogate modeling of soil water flow. DPL-S discretizes the SM state evolution into multiple sub-process effects (e.g., gravity, matric potential) at the intuitive physics level and abstracts them into format-specific and learnable tensors. By cascading state-action matrices in a differentiable end-to-end framework and enforcing penalties for physical inconsistencies, DPL-S enables a profound understanding of physical functions and scenes of soil water flow. Comprehensive numerical experiments including layered soil conditions and tests with in situ observations, demonstrate that it achieves reliable SM profile reconstruction with predictive performance comparable to the state-of-the-art DL model on supervised items. The internal inference of DPL-S is fully transparent and the tensor representations achieve strict physical realism under limited water content supervision, thus enabling continuous predictions like physical models during the testing period. The model’s flexibility, generalization, noise resistance, and large-sample diverse data synergies are also evaluated. This work represents a solid step toward learning hydrophysical processes from large data sets.

2025 Science (with a capital S!) breakthrough of the year

Does Science with a capital S speak for science? I don’t know, science, or nature or Nature might have something to say about that. Small-s science, after all, is just a way of asking questions and trying to strengthen our confidence in what we think we know about nature. Despite all that, the magazine/publishing conglomerate known as Science nominates candidates for scientific breakthrough of the year and then chooses one. This year’s winner is renewable energy.

This year, renewables surpassed coal as a source of electricity worldwide, and solar and wind energy grew fast enough to cover the entire increase in global electricity use from January to June, according to energy think tank Ember. In September, Chinese President Xi Jinping declared at the United Nations that his country will cut its carbon emissions by as much as 10% in a decade, not by using less energy, but by doubling down on wind and solar. And solar panel imports in Africa and South Asia have soared, as people in those regions realized rooftop solar can cheaply power lights, cellphones, and fans. To many, the continued growth of renewables now seems unstoppable—a prospect that has led Science to name the renewable energy surge its 2025 Breakthrough of the Year…

China’s mighty industrial engine is the driver. After years of patiently nurturing the sector through subsidies, China now dominates global production of renewable energy technologies. It makes 80% of the world’s solar cells, 70% of its wind turbines, and 70% of its lithium batteries, at prices no competitor can match.

The article makes the point that this progress is not really a technological breakthrough, but rather a successful scaling up of technology invented during the space race half a century ago. Materials science does offer some possibilities for breakthroughs on the near horizon:

Solar cells today are made of crystalline silicon, but another kind of crystal, perovskites, can be layered in tandem with silicon to make cells that gain efficiency by capturing more colors of light. Material advances are enabling wind turbine blades to get longer and harvest more energy, while designs for floating turbines could vastly expand the offshore areas in which they could be deployed. And the giant lithium-ion batteries now used to store energy when sunshine and wind falter could one day give way to other chemistries. Vanadium flow batteries and sodium batteries could be cheaper; zinc-air batteries could hold far more energy.

And there you go – an agenda for research and development that the U.S. could get behind, or better yet, cooperate internationally on a win-win basis.

Meanwhile the nominees that were not chosen were:

  • Gene-editing to cure rare diseases in human babies and adults
  • New antibiotics effective against antibiotic-resistant gonorrhea, which continues to evolve
  • A breakthrough in understanding how cancer can spread through the nervous system
  • Advances in telescopes
  • DNA reconstruction of early humans
  • Large language models conducting math and scientific experiments on their own – In 2025 this was done with thorny math problems, chemical and drug development. The article notes that AI agents did not really live up to their hype overall in 2025.
  • Stuff involving subatomic particles. Honestly, this stuff is interesting but it’s hard for us normals to draw straight lines to how it might eventually affect our daily lives. Of course this doesn’t mean it won’t, it just means a lot of twists and turns as it works its way through the worlds of science and technology over time.
  • Genetically engineered organs grown in pigs and transplanted to people (successfully, at least for a period of months which seems to be much longer than these particular people were expected to live without the experimental transplants.). Are these pig organs or human organs grown in pigs? At some point it doesn’t matter.
  • Advances in heat-resistant rice

The article makes a parting shot at the U.S. government under Trump, for just intentionally shooting our entire scientific development pipeline in the foot. These were not the actions of a patriot, if I need to remind anyone.

BBC: 25 most important scientific ideas of the 21st century

BBC has a list called The 25 most powerful ideas of the 21st century (so far), picked by the world’s top thinkers. They don’t spell out science or technology in the title, but I don’t see any grand philosophical or literary analysis here. It’s not exactly clear if the list is in any order, other than maybe grouped loosely by topic. I’m just going to list a few I found interesting below, in categories:

  • Medicine: stem cells that don’t come from babies, mRNA vaccines, genome sequencing, a cure for HIV*, the HPV vaccine, contraception apps [what we used to call “the rhythm method and were cautioned not to use, but the apps now make it accurate], tissue engineering [this is growing body parts from a sample of human DNA for implant back into that same person – the article says ears, trachea, and bone have been used in patients, while kidneys and hearts are still at the research stage], psychedelic therapy
  • Environment: global warming and continuing carbon emissions, attribution analysis
  • (Information) Technology: large language models, robots that can do chemistry experiments
  • (Other) Technology: self-repairing materials
  • Physics/Cosmology: dark matter, the Higgs boson, the James Webb telescope, exoplanets, gravitational waves

* The HIV cure deserves some extra discussion. HIV can be cured, at least in some people sometimes, by transplanting bone marrow from a naturally HIV-resistant person. A bone marrow transplant is such a big deal that it would not be ethical to do it for people whose only problem is HIV(!) because other effective treatments are available. It is done for people with terminal leukemia when no other treatments are available. A few of these people have HIV, and it has been shown that their HIV can be cured. So we need to keep working on applying some of the other technologies to an HIV vaccine and/or cure.

I want to just briefly talk about the contraceptive apps. I might have heard about that but didn’t realize it had been so rigorously studied and FDA-approved. It seems so simple and yet a breakthrough, which I find heartening. I find this heartening because I would like to see our society eventually move on from the abortion debate, and the way to move on in my view is to improve technology, access and knowledge about birth control while reducing stigma. This seems to me to accomplish all those objectives without a major scientific breakthrough being required. (I am under no illusions about the politics – if technology solves an issue, people who need an issue to suit their political purposes will find or manufacture another issue.)

ASPI Critical Technology Tracker

Something called the Australian Strategic Policy Institute tracks and forecasts which countries in the world are leading on what it considers the most critical technologies. Their definition of critical seems to be mostly technologies with military applications: “defence, space, energy, the environment, artificial intelligence (AI), biotechnology, robotics, cyber, computing, advanced materials and key quantum technology areas”. And their metrics seem to be based largely on number of scientific publications and patents. This approach can be critiqued, but nonetheless the results are interesting and striking.

These new results reveal the stunning shift in research leadership over the past two decades towards large economies in the Indo-Pacific, led by China’s exceptional gains. The US led in 60 of 64 technologies in the five years from 2003 to 2007, but in the most recent five years (2019–2023) is leading in seven. China led in just three of 64 technologies in 2003–20074 but is now the lead country in 57 of 64 technologies in 2019–2023, increasing its lead from our rankings last year (2018–2022), where it was leading in 52 technologies…

China’s new gains have occurred in quantum sensors, high-performance computing, gravitational sensors, space launch and advanced integrated circuit design and fabrication (semiconductor chip making). The US leads in quantum computing, vaccines and medical countermeasures, nuclear medicine and radiotherapy, small satellites, atomic clocks, genetic engineering and natural language processing.

Building technological capability requires a sustained investment in, and an accumulation of, scientific knowledge, talent and high-performing institutions that can’t be acquired through only short-term or ad hoc investments.8 Reactive policies by new governments and the sugar hit of immediate budget savings must be balanced against the cost of losing the advantage gained from decades of investment and strategic planning. While China continues to extend its lead, it’s important for other states to take stock of their historical, combined and complementary strengths in all key critical technology areas.

I suppose the not-so-hidden agenda here is to get the Australian and other “western” governments to invest more in R&D long-term. That is something I would support. I would like to think that technological progress is not just a competition between nation-states but a shared project of our species and civilization. Utopian, I suppose.

Anyway – scientific publications and patents. I don’t think these are perfect measures of scientific or technological progress. Doubling these metrics will not mean that progress has doubled, but rather there must be some diminishing return. Once metrics like these are established, people are going to game the metrics to some extent rather than try to measure the underlying thing, which in this case is scientific and technological progress.

Do I have a better suggestion? Not really – well, I suppose total factor productivity is the most accepted metric of technological progress as far as I know. The holy grail would be to understand exactly how much and what types of R&D investments will maximize it over long periods of time. I am sure there are past and future Nobel laureates working on this problem, but if they have solved in conclusively I have not heard about it.

All that said, there is no excuse for the U.S. to be failing to invest in R&D. We need to ramp it up, and keep it up long term. But there is also an opportunity cost when the fire hose is focused on the military-industrial complex (not to mention the existential risks created for us and all humanity – do these alone outweigh the idea of ever winning the “competition” for dominance in horrible weapons?). Peaceful technologies that could improve human lives and our shared environment will not develop as fast as they could. And finally, to be a broken record, if we ever figure out the secret sauce to ramp up scientific and technological progress, the right thing to do is capture that value added to the economy and redirect it to improve the vast majority of human lives, protect the environment, and manage the risks we face, including risks created by the technologies themselves.

AI and protein research

Here is a story in MIT News about AI doing experiments on proteins, with drug development and gene therapy implications. This seems like the clearest application of AI at the moment – anything where there is a formula to be figured out and a large number of combinations to be tried. I can definitely see this accelerating scientific and technological progress, although the efficiency to me seems to be more in the “automation” part than the “intelligence” part.

January 2024 in Review

Most frightening and/or depressing story: 2023 was “a year of war“, and so far 2024 is not looking better. Those diplomatic grand bargains you always hear about seem to be getting less grand. And the drumbeat for a U.S. attack on Iran got louder.

Most hopeful story: According to Bill Gates, some bright spots in the world today include gains in administering vaccines to children around the world, a shift toward greater public acceptance of nuclear power, and maybe getting a bit closer to the dream of fusion power. He pontificates about AI, and my personal sense is it is still too soon, but AI does hold some promise for speeding up scientific progress.

Most interesting story, that was not particularly frightening or hopeful, or perhaps was a mixture of both: The return of super-sonic commercial flight is inching closer.

October 2021 in Review

Most frightening and/or depressing story: The technology (sometimes called “gain of function“) to make something like Covid-19 or something much worse in a laboratory clearly exists right now, and barriers to doing that are much lower than other types of weapons. Also, because I just couldn’t choose this month, asteroids can sneak up on us.

Most hopeful story: The situation with fish and overfishing is actually much better than I thought.

Most interesting story, that was not particularly frightening or hopeful, or perhaps was a mixture of both: I thought about how to accelerate scientific progress: “[F]irst a round of automated numerical/computational experiments on a huge number of permutations, then a round of automated physical experiments on a subset of promising alternatives, then rounds of human-guided and/or human-performed experiments on additional subsets until you hone in on a new solution… [C]ommit resources and brains to making additional passes through the dustbin of rejected results periodically…” and finally “educating the next generation of brains now so they are online 20 years from now when you need them to take over.” Easy, right?