Author Archive: Stephen Green

21ST CENTURY WARFARE: CENTCOM using underwater drones to clear mines in the Strait of Hormuz. “Iran is estimated to have between 2,000 and 6,000 naval mines, and upward of 80-90% of its small boats and mine layers, making it possible to lay hundreds of mines in the waterway. On Saturday, US officials said that Iran reportedly lost track of the locations of mines deployed in the Strait of Hormuz and has no clear idea of where all the mines were placed.”

THE LITTLE COUNTRY THAT COULD:

Plus, a reminder that US aid to Israel is only $3.8 billion annually, and almost all of that comes right back here as arms purchases.

KRUISER’S MORNING BRIEFING: Abigail Spanberger’s Virginia Is a Creepy, Commie Place. “Democratic Governor Abigail Spanberger is leading an unhinged leftist revival in Virginia that has been rather horrifying in its swiftness. Democrats have complete control of the government now and the wraith-like Spanberger has them all whipped into a frenzy and ready to charge off of the radical progressive cliff as quickly as they can.”

“INCOMPETENT TO STAND TRIAL” SHOULD ALSO MEAN “UNABLE TO LEAVE THE ASYLUM”:

Unleash crazy people on the public, expect crazy results.

HMM: Is Anthropic ‘nerfing’ Claude? Users increasingly report performance degradation as leaders push back.

A growing number of developers and AI power users are taking to social media to accuse Anthropic of degrading the performance of Claude Opus 4.6 and Claude Code — intentionally or as an outcome of compute limits — arguing that the company’s flagship coding model feels less capable, less reliable and more wasteful with tokens than it did just weeks ago.

The complaints have spread quickly on Github, X and Reddit over the past several weeks, with several high-reach posts alleging that Claude has become worse at sustained reasoning, more likely to abandon tasks midway through, and more prone to hallucinations or contradictions.

Some users have framed the issue as “AI shrinkflation” — the idea that customers are paying the same price for a weaker product.

Others have gone further, suggesting Anthropic may be throttling or otherwise tuning Claude downward during periods of heavy demand.

Those claims remain unproven, and Anthropic employees have publicly denied that the company degrades models to manage capacity. At the same time, Anthropic has acknowledged real changes to usage limits and reasoning defaults in recent weeks, which has made the broader debate more combustible.

VentureBeat has reached out to Anthropic for further clarification on the recent accusations, including whether any recent changes to reasoning defaults, context handling, throttling behavior, inference parameters or benchmark methodology could help explain the spike in complaints.

Anthropic lost maybe as much as $6 billion last year, and is expected to lose another $8-10 billion this year. Revenue is growing, but infrastructure costs are growing even faster.

Maybe something had to give.

JAW, JAW: Trump tells The Post US-Iran talks ‘could be happening over next two days.’

Trump did not say who would represent the US in a potential second round of talks, but confirmed he would not take part.

The president also indicated he was not pleased with reports that the US had asked Iran to suspend its uranium enrichment program for at least two decades during this past weekend’s unsuccessful talks.

“I’ve been saying they can’t have nuclear weapons” he said, “so I don’t like the 20 years.”

Asked about proponents suggesting a moratorium may encourage Iran make an agreement, Trump answered: “I don’t want them [Iran] to feel like they have a win.”

Not sure what a second round of talks might accomplish that the first round didn’t, particularly since the 20-year deal seems to be off the table.

Besides, all the really interesting talk seems to be happening in Iran already:

Iran doesn’t seem to be playing its A game, but to be fair, the US and IDF killed most of its A players.

UPDATE (FROM GLENN):

DISPATCHES FROM THE BLUE ZONES:

THAT EXPLAINS IT:

TO BE FAIR, MILEI IS RIGHT ABOUT PRETTY MUCH EVERYTHING:

BE TOLERANT, THEY TOLD ME, OR ELSE:

JOANNE JACOBS: It’s a marathon, not a miracle.

As founder of the education nonprofit Mississippi First, Canter saw her home state start getting serious about meeting higher expectations in 2008. That change included state power to take over low-performing districts, A-F grades for schools and districts based on student achievement and challenging new learning standards.

The state’s literacy law passed in 2013: Schools screen students’ reading skills three times a year and report progress to parents. Students can’t read adequately by the end of third grade are held back a year. The law meant “everyone in the system would be in a hellfire hurry to teach children to read,” writes Canter. “No one wanted children to fail.”

Accountability works — who’d have thought?

IF BY WEIRD YOU MEAN BUSINESS AS USUAL:

IF ONLY EUROPE WOULD MAKE CRUSADES GREAT AGAIN:

Why, it’s almost as if…:

Islam by Design

Update: Sorry, the embed is broken for some reason and just noticed. Hope the screencap helps.

NOTHING TO SEE HERE, MOVE ALONG: