Michael Munger performs an autopsy: Chatbots Killed the Academic Star.
Having been an editor on three journals, I can say that the most difficult thing a journal must do is find referees. So I was thinking last week of how nice it would be to have a “refbot” (software that was trained on what made for a good, publishable article) to use as a referee. And then I suddenly realized that I was contemplating a new kind of singularity.
- AI chatbots start writing papers and sending them to journals (This is pretty much happening now, folks). At first, the chatbots would be used by specific researchers to assist in writing a paper that was going to be written anyway. But there’s a lot of downtime, and writing is easy.
- Once sent to the journal, refbots, “trained” for the specific journal by analyzing the corpus of “good” (previously published) academic papers in that journal, perhaps leavened by other academic work the editors aspire to have their journal mimic, will evaluate the submissions. The refbots will investigate contradictions, look for consensus, and check references. (That may sound pretty superficial, but that would be better than about 2/3 of the human referees journals can actually find, and it would be fast. On the other hand, it appears that perhaps the tech is not quite there yet.)
- The feedback loop is then closed by the next generation of chatbots scanning the published literature and deciding what is important, citing that work in the next round of published articles. The articles that attract the most citations from the next generation of chatbot authors, and the next, will get higher status in search engines that return the “best” articles for human researchers to use, after the selection process has culled the dross.
I'm not sure how much of Munger's article is tongue-in-cheek. See what you think.
Jacob Sullum adds a few more adjectives to the usual "demented" and "dishonest": Biden Looks Careless, Shady, and Hypocritical After the Revelations About His Handling of Classified Material.
In addition to the "small number" of classified documents in President Joe Biden's former think tank office, it turns out, he had a "small number" in the garage of his house in Wilmington, Delaware, plus one more in a room adjacent to the garage. These were Obama administration records that Biden came across during his time as vice president, and they were definitely not supposed to be in those locations. What had initially seemed like a single lapse now looks like a pattern of carelessness, which creates several problems for Biden and the Justice Department.
First, Biden is no longer in a position to criticize Donald Trump's "totally irresponsible" handling of sensitive material that he retained when he left office. Second, the delay in acknowledging Biden's retention of classified records and obfuscation of its scope look like blatant attempts to minimize the political fallout. Third, a criminal prosecution of Trump for his handling of the government documents he took to Mar-a-Lago, which was always an iffy proposition, now seems doomed for political as well as legal reasons.
Well, it's not as if there aren't other lying politicians in the news.
Eric Boehm makes a point that's obvious to any sentient being: Cutting Government Back to Last Year's Size Wouldn't Be 'Impossible' or 'Severe'.
As part of a deal struck between House Speaker Kevin McCarthy (R–Calif.) and the fractious House Freedom Caucus, Republicans in Congress have pledged to return federal discretionary spending to 2022 levels.
As a practical matter, that would require cutting about $130 billion out of the federal budget next year. But it's probably more useful to think about the maneuver as an attempt to rescind the spending increases included in the omnibus bill that Congress rushed to pass in the final weeks of last year. That bill set spending levels for the 2023 fiscal year, so promising to return to the 2022 spending level amounts to a promise to undo that omnibus bill and not replace it with more spending hikes.
In a more normal place, this would be described as what it is: a promise to hold government spending level. The federal government spent about $1.7 trillion on discretionary programs in 2022, and Republicans are saying they'd like to spend the same amount next year.
Eric rattles off some of the news stories that show that Washington D. C. is far from a "more normal place".
As noted yesterday, American airlines and their regulators do a pretty good job keeping us from dying in spectacular crashes. On the other hand… Flight Chaos Demonstrates Need for Systemic Changes in Air Traffic Control Policy. Or so says Colin Grabow at Cato:
Air travel in the United States was thrown into chaos earlier this week when a key system used by the Federal Aviation Administration (FAA) went down, forcing over 10,000 flights to be delayed and at least 1,300 canceled. While a damaged database file may have been the proximate cause of this upheaval, the episode appears yet another indication of systemic flaws in U.S. air traffic control policy.
Technological deficiencies in U.S. air traffic control operations are long-standing. Indeed, Congress mandated twenty years ago that the FAA establish a plan for implementing the modernized Next Generation Air Transportation System (NextGen) by 2025 in order to improve matters. The FAA’s technological woes, however, are unlikely to go away anytime soon. A 2021 Department of Transportation Office of Inspector General report noted that the agency has “struggled to integrate key NextGen technologies and capabilities”—a finding consistent with other reports on the topic—and the particular system that failed this week apparently won’t be updated for another six years.
Interesting fact: other countries have moved to privatize their air control systems with good results. So far that's been a non-starter in the US. Sad!