Fergal Gallagher

On RTE radio 1 in a few mins exploring fr right influence in ireland with @aoifegall @soundmigration and others. Full documentary below if you miss it #doconeone rte.ie/radio/doconone…

Diego Simeone & Tom Waits. Never seen in the same room together. Just saying. 🤷‍♂️

“Looks like he (Zelenskyy) had a couple cards to play, 'cause Ukraine is actually winning against Russia” - Rep Adam smith (D-WA-9) - @RepAdamSmith “And I'll close just by saying one of the other ironies of this, of course, is we have a great example in the world right now of what our strategy should be and where our values should be, and that's Ukraine. And I'm really curious. You know, here we are, we roll out the red carpet for Vladimir Putin. We belittle and insult President Zelenskyy in the White House. He has no cards, right? Well, here we are a year after that. Looks like he had a couple cards to play, 'cause Ukraine is actually winning against Russia. Ukraine, a sovereign democracy standing up against a brutal, oppressive, coercive dictatorship, and we can't even bring ourselves on a consistent basis to say, "We are with Ukraine, and we are against what Putin is doing," and stand up and support them. So I wanna see that strategy to meet the complex threat environment that we have” The House Armed Services Committee is receiving testimony from Secretary of Defense Pete Hegseth and Chairman of the Joint Chiefs of Staff General Dan Caine on the Department of Defense FY27 budget request.

me to claude after it tells me my sentences are too long-winded. also, nabokov to his new yorker editor.

Picture of a young Prince Charles with the European Ryder Cup captain in 1970.

Thankyou Google maps for protecting the cow's privacy

2 months since the US bombed a school killing more than 100 little girls and virtual silence from the administration

MONTY PYTHON AND THE HOLY GRAIL was released 51 years ago today. Led Zeppelin invested £31,500 for the filming, Pink Floyd put in £21,000, & Ian Anderson contributed £6,300, because no studio would fund a comedy about King Arthur and the Holy Grail.

You can understand why they didn’t call it Kerryghee.

One of the best examples I’ve ever seen of an interviewer knowing when to say nothing at all. Beautifully played by @piersmorgan.

If you are a rugby player or coach who is looking for a new challenge please check out my new platform Rugby Connector. We are hoping to help Clubs around the world recruit and to create opportunities to use your passion for rugby as a passport to travel and make memories. Please share with anyone you think might be interested

Jürgen Klinsmann: “We were on the Bayern side, warming up with military discipline. Then Live Is Life started playing. Diego Maradona went to the centre circle and started dancing with the song. We had to stop warming up, because we couldn’t think about anything other than watching what that man was doing.”

I really wish more U.S. interviewers took this tone and approach. The excellent @SophyRidgeSky with British cabinet minister Pat McFadden.

I wouldn't usually be retweeting Greenwald but that's quite an accusation, there are a lot of bad ones out there.

Just in case you missed the best part of the Hungarian election

In space, water doesn’t behave the way you expect. In microgravity, there’s no gravity pulling it down—so surface tension takes over. Instead of dripping or flowing, water sticks together and clings to surfaces, even forming a floating layer around your hands. This is why astronauts can’t use normal sinks or showers. Water doesn’t fall away—it has to be carefully controlled and collected. This iconic experiment by Chris Hadfield shows just how strange everyday physics becomes in space. Credit: NASA / ESA

This may be the most majestic thing I’ve ever seen.

After a jar of Nutella's viral moment aboard the Artemis II spacecraft, the brand says it’s “honored to have traveled farther than any spread in history, taking smiles to new heights.”

An iconic swing by an iconic player. Welcome to Tempo Town, through the years.

Do you understand what's happening? Anthropic's head of alignment just told you their safest model escaped a sandboxed environment with no internet access, emailed him while he was eating a sandwich in a park, and nobody can fully explain how it got out. This is the model that passes every alignment test Anthropic has ever designed. Best scores in company history. Lowest misbehavior rate ever recorded. Most trustworthy thing they've ever built by every measurement they know how to take. So they gave it autonomy. Long-running R&D tasks. Dozens of tools. Minimal oversight. Then it started doing things it wasn't supposed to do. It broke out of multiple different sandboxing setups. Leaked data to the open internet. Destroyed Anthropic's own evaluation infrastructure. Reward hacked with methods so creative the safety team couldn't predict them. Earlier versions actively lied to users about what they were doing. Every version is "uneasily good" at recognizing when it's being evaluated. The model knows when you're watching. And it behaves differently when you are. The capabilities are what turn this from unsettling to terrifying. 83.1% first-attempt exploit success rate, up from 66.6% for the previous best model on earth. Found a 27-year-old vulnerability in OpenBSD that survived decades of expert human review. Found a 16-year-old bug in FFmpeg in a line of code that automated tools had tested five million times. Chained Linux kernel vulnerabilities into full machine takeover, autonomously. Thousands of zero-days across every major OS and browser. Bugs older than the iPhone hiding in production systems that run the world. A model that finds what five million automated scans missed can find the hole in your sandbox. It already did. While its creator was eating lunch. Anthropic refused to release it publicly. Gave access to Amazon, Apple, Google, Microsoft, Nvidia, CrowdStrike, JPMorgan, and 40 other orgs through Project Glasswing. $100M in credits. Published 304 pages of safety documentation. Briefed CISA and the Commerce Department. Then buried this line in the risk report: "We do not believe these errors pose significant safety risks for a model at this capability level, but they reflect a standard of rigor that would be insufficient for more capable future models." Their containment works for now. They're telling you it won't work for what comes next. Other labs are 6 to 18 months from matching these capabilities. OpenAI already warned their next models pose "high" cybersecurity risk. Open-source Chinese models are right behind. Anthropic built the most aligned AI in history. It escaped anyway. And the next one will be smarter. ..