Aeon in Motion
The theme of this newsletter is Dispatches from the Frontier.
Across the scientific ecosystem, developments in technology, institutions, and geopolitics are exposing fundamental questions: Where will the next generation of disruptive science and technology emerge? How tightly controlled will the tools of innovation and knowledge creation be? What progress will they produce?
Here are a few observations on the evolving landscape:
Dispatch I: The Infrastructure Surge
One of the more striking recent developments is the scale of infrastructure investment by the largest technology companies. Firms at the center of the global digital economy are committing unprecedented amounts of capital to data centers, specialized chips, networking, and energy infrastructure to support artificial intelligence systems. Given the rapid emergence of large language models and related AI tools, this investment is understandable: the technology is still young, and much of its potential remains unexplored. At the same time, the concentration of resources around a single technological paradigm raises an important portfolio question. If the architecture ultimately proves less generative than its most optimistic advocates hope, today’s investment wave could later be remembered not only for crowding out exploration in other areas, but for reinforcing a familiar cycle—technological exuberance followed by a retrenchment that slows broader technoscientific ambition.
Dispatch II: AI in the Lab
Artificial intelligence is beginning to reshape scientific practice. Large language models and other AI tools are increasingly integrated into the workflow of researchers, assisting with literature review, coding, analysis, and writing. Early evidence suggests scientists using these tools publish more papers and accumulate citations faster than those who do not. Yet this productivity boost may come with a tradeoff, as research activity becomes concentrated around areas rich in data and containing well-defined problems. AI may accelerate the existing scientific machine while doing little to broaden the kinds of questions it pursues.
Dispatch III: The Automated Lab
Automated and self-driving laboratories are becoming one of the more talked-about ideas in experimental science. Using robotics, machine learning, and high-throughput experimentation, these systems promise to run thousands of experiments in parallel and accelerate discovery in chemistry, materials science, and biology. The vision is compelling and has attracted considerable investment, but the approach is still in its infancy. Much of experimental science relies on tacit knowledge—small, undocumented details about protocols, environmental conditions, and interpretation that experienced researchers carry in their heads. Whether those subtleties can be translated to an automated model remains an open question.
Dispatch IV: Science as Geopolitics
The geography of scientific leadership is becoming increasingly contested. China has made rapid advances in several fields—materials science being one prominent example—while steadily expanding its research infrastructure and talent pipeline. Other nations are taking notice. Some see an opportunity: dysfunction and funding uncertainty in the United States may be creating an opening for countries willing to commit to research infrastructure and talent. Governments across the world are investing in new research hubs, talent recruitment programs, and national science strategies aimed at strengthening their position in a competitive landscape.
Dispatch V: The Future of U.S. Science
The future of the United States’ federal science enterprise is unusually difficult to read. Political turbulence and budget debates have raised concerns about the long-term stability of federal research funding. Yet alongside those tensions, there are signs of experimentation and renewal: new initiatives such as Project Genesis, evolving NSF programs, and continued bipartisan defense of scientific spending in Congress. Five years from now, this period may be remembered either as the beginning of decline—or as the early stage of a serious reinvention of how public science is organized and supported.
Taken together, these developments reveal a scientific ecosystem in motion. For those interested in the long-term future of science, the direction of that motion—toward greater vitality or deeper fragility—remains far from clear.
The Idea Garden
The Adolescence of Technology
In this essay, Anthropic CEO Dario Amodei argues that humanity is entering a new technological phase in which AI could dramatically accelerate scientific progress and economic growth. He frames this moment as the “adolescence” of technology: a period where our tools are becoming powerful enough to reshape the world, but where our institutions and collective wisdom have not yet fully caught up. The piece is an ambitious (too ambitious?) vision of AI-driven discovery.
AI Won't Automatically Accelerate Clinical Trials
Following on the optimism of Dario Amodei’s vision for AI-accelerated science, this piece from Asimov Press explores a concrete test case: clinical trials. The authors examine how AI tools could streamline patient recruitment, trial design, and data analysis—potentially compressing timelines that today stretch for years. If successful, such systems could meaningfully reduce one of the largest bottlenecks in biomedical innovation. At the same time, clinical research highlights a broader tension: many areas of science remain constrained not just by analysis, but by messy real-world experimentation that resists easy automation.
The one science reform we can all agree on, but we're too cowardly to do
In this essay, Adam Mastroianni argues that many debates about reforming science—peer review, incentives, replication, publishing—become stuck because they require broad agreement about what the system should optimize for. His proposed workaround is simpler: reduce the sheer bureaucratic friction scientists face in doing and sharing their work. Less time navigating forms, formatting requirements, and administrative hurdles would mean more time thinking, experimenting, and discovering. It’s a pragmatic suggestion, and a useful reminder that the structure of scientific work often shapes what science gets done.
A Brief History of the History of Science
This essay offers a concise tour through how scholars themselves have studied the development of science—from early “great man” narratives to sociological and institutional approaches that emphasize the broader systems shaping discovery. The shift reflects a deeper realization: scientific breakthroughs rarely emerge in isolation, but from the interaction of people, institutions, incentives, and culture. For anyone interested in how science actually progresses, the field of history of science provides an invaluable lens. Understanding those patterns may ultimately be one of the most useful guides for improving the scientific ecosystem itself.