More from Asterisk
Building a high-speed rail between Los Angeles and San Francisco was never going to be easy — but the critics who write it off are missing the real source of the project’s struggles.
A conversation about the factors that might slow down the pace of AI development, what could happen next, and whether we’ll be able to see it coming.
Kelsey Piper and Jasmine Sun talk about microschools, whether localism is the enemy of Abundance, and why Chinese bureaucrats are like Growth PMs.
USAID has been slashed, and it is unclear what shape its predecessor will take. How might American foreign assistance be restructured to maintain critical functions? And how should we think about its future?
More in science
Naturally, the early history of space exploration is filled with firsts. Just six decades ago at this time, the world watched as NASA’s Mariner 4 spacecraft […]
The US House and Senate appropriations subcommittees have now completed their markups on the bills relevant to the FY26 appropriations for NSF, NASA, and NIST. The AAAS has an interactive dashboard with current information here if you want to click and look at all the science-related agencies. Other agencies still need to go through the Senate subcommittees. Just a reminder of how this is supposed to work. The House and Senate mark up their own versions of the detailed appropriations bills. In principle these are passed by each chamber (with the Senate versions for practical purposes requiring 60/100 votes of support because of the filibuster). Then a conference committee hashes out the differences between the bills, and the conference version of the bills is then voted on by each chamber (again, needing 60/100 votes to pass in the Senate). Finally, the president signs the spending bills. In the fantasy land of Schoolhouse Rock, which largely described events until the 1990s, these annual spending bills are supposed to be passed in time for the start of the new fiscal year on October 1. In practice, Congress has been deeply dysfunctional for years, and there have been a lot of continuing resolutions, late budgets, and mammoth omnibus spending bills. To summarize: NSF - House recommendation = $6.997B (a 20.7% cut from FY25), Senate = $9B (a 2% increase from FY25). These are in sharp contrast to the presidential budget request (PBR) of a 55.8% cut. NASA - House = flat from FY25, Senate = $24.9B (0.2% increase). NIST - House = $1.28B (10.6% increase from FY25), Senate = $1.6B (38.3% increase from FY25) NOAA - House = $5.7B (28.3% increase from FY25), Senate = $6.1B (36.3% increase from FY25) DOE has gone through the House, where the Office of Science is recommending a 1.9% increase, in contrast to a 13.9% cut in the PBR. If you are eligible and able to do so, please keep pushing. As I wrote a few days ago, this is a long-term project, since appropriations happen every year. As long as you're making your opinions known, it's good to push on representatives and senators that they need to hold the agency leadership accountable to actually spend what congress appropriates. A science post soon....
Fundamental technique lets researchers use a big, expensive “teacher” model to train a “student” model for less. The post How Distillation Makes AI Models Smaller and Cheaper first appeared on Quanta Magazine
When Europeans arrived to the Pacific Northwest, they spread smallpox that devastated the Indigenous people, plundered stocks of salmon and herring, hunted down deer and other game, and built sprawling cities and ports. New research tallies the profound impact on wildlife. Read more on E360 →
A team of mathematicians based in Vienna is developing tools to extend the scope of general relativity. The post A New Geometry for Einstein’s Theory of Relativity first appeared on Quanta Magazine