After weeks of being shut down due to safety concerns (via questionable export controls), Anthropic’s Claude Fable is once again accessible to the public with new guardrails that limit its capabilities and revert to inferior models more frequently. OpenAI’s ChatGPT‑5.6 has had a delayed rollout to fulfill a reported request from the White House. These models have been evaluated, withheld from the market, and re-released. But no one knows what made them too dangerous to release, and no one knows what would make them safe enough to release. The obscure and seemingly arbitrary standards of the current evaluation process established by the White House’s executive order on Frontier Artificial Intelligence (AI) show that, while it did not formally introduce an “FDA for AI,” it might have done so in practice. In fact, this regime shares most of the common pitfalls of a vetting regime and, in some cases, might worsen them. 

Read the rest of this post →

Last night, President Trump gave a speech, mostly on voter fraud and elections, that many television networks, including ABC and NBC, chose not to air. 

As a result, President Trump accused ABC and NBC in his speech of being part of the fraud, a fraud that “should mean a revocation of their licenses; they use our public, multi-billion dollar in value airwaves for absolutely no money; they pay nothing.” 

Cato scholars have been clear that the government does not and should not have the power to censor or compel media organizations in their reporting or programming. While a presidential address is often newsworthy, media organizations make their own editorial decisions that are protected by the First Amendment. In 2022, most networks, including ABC and NBC, refused to cover a prime-time address by President Biden in which he described Republicans as threats to democracy. Similar to last night’s speech, they did so because they felt the topic was political in nature and right before a midterm election. Just as no government has the right to demand what newspapers or cable news covers or does not cover, no government should have that power over broadcast media. 

Read the rest of this post →

Elizabeth Watkins, director of Ascend Micro School in Colorado Springs, didn’t set out to run a school. She set out to find the right one for her son.

The oldest of five kids, Elizabeth grew up homeschooled in northern Maine and credits that upbringing with giving the close family bonds she still values today. She always assumed she’d homeschool her own children. What she didn’t anticipate was how hard it would be.

Her oldest was “very independent, very wanting to do things his own way,” she says. He also had some learning challenges that meant his way of learning was “180 degrees different than the way I learned,” she adds. After two years of homeschooling him, which left them both in tears, Elizabeth enrolled him in public school. That wasn’t quite right either.

Read the rest of this post →

Senator Elizabeth Warren claims that the Trump administration’s cutbacks at the Consumer Financial Protection Bureau (CFPB) cost consumers $26.5 billion. But the method she used to calculate this scary number is hopelessly oversimplified. Regulation has costs, and those costs need to be balanced against benefits. Indeed, the Dodd-Frank Act calls for the bureau to consider the costs and benefits of its actions, especially how these actions affect consumers’ access to credit. Senator Warren seems to have forgotten this basic instruction.

Under the Biden administration, the CFPB capped credit-card late fees and limited bank overdraft fees. Senator Warren claims that these rules would have saved consumers over $20 billion. But both the caps on late fees and on overdraft fees would also have harmed consumers. Late fees discourage consumers from making late payments and overspending. And if fees are capped, card issuers are likely to limit the costs of late payments by lending less to consumers in need or charging higher fees elsewhere—a well-known effect of regulating credit card charges. The CFPB’s limit on overdraft fees would have saved people who bounce checks some money in the short run—but other consumers would lose out, as overdraft fee caps would lead banks to reduce overdraft coverage and raise minimum account balances. 

Read the rest of this post →

Today, the Trump administration made public two new rules limiting legal immigration: one, a massive overhaul of the international student visa system; the other, a rewrite of the public charge ground of inadmissibility that will primarily hit relatives of US citizens—overwhelmingly their spouses and minor children. 

As I have detailed, Trump had already cut legal immigration far more than illegal immigration—2.5 times as much. These two rules will exacerbate that trend. Neither targets anyone who crossed the border illegally. Both target people who did exactly what the law asked of them.

The Student Rule: Banned from Changing Your Mind 

Previously, international students on F visas, exchange visitors on J visas, and foreign media on I visas were granted admission for their “duration of status”—that is, for as long as they were in status or following the rules of the visa programs. DHS’s final rule replaces that with a fixed period of entry of no more than four years (for international students and exchange visitors). But it does far more than end open-ended admission. Littered throughout it is a set of unprecedented restrictions with no statutory basis. 

• Read the rest of this post →

Last night during his national address, President Trump claimed that key documents he ordered declassified demonstrate that the US Intelligence Community (IC) withheld or otherwise soft-pedaled evidence of foreign election interference or voting machine vulnerabilities. The documents at issue—released in a process that can be charitably described as irregular at best—tell a different story.

One very recent CIA document (dated June 29, 2026) had this to say about allegations of Venezuelan government efforts to manipulate electronic voting systems:

Read the rest of this post →

A recent study points out the dearth of historical research on the macroeconomic effects of tariffs, especially as a tool to analyze modern-day tariffs. The study

addressed this challenge by analyzing all major US tariff rate changes from 1840 to 2024. Drawing on historical research, congressional records, and statutes, we identified … 21 tariff rate changes, which we used to examine the macroeconomic effects of tariffs.

These examples

reveal that increasing tariff rates contracted the US economy. … [T]ariff increases did not shield domestic industry despite their protective intent. Additionally, trade contracted markedly. 

Read the rest of this post →