Mike Nicholson

Anthropic is rejecting the Pentagon’s latest offer to change their contract, saying the changes do not satisfy the company’s concerns that AI could be used for mass surveillance or in fully autonomous weapons. https://cnn.it/4tWJ0wD

The political effects of X's feed algorithm https://doi.org/10.1038/s41586-026-10098-2 Received: 16 December 2024 Accepted: 4 January 2026 Published online: 18 February 2026 Open access • Check for updates Germain Gauthier,5, Roland Hodler?5, Philine Widmer35 & Ekaterina Zhuravskaya3,4,5 m Feed algorithms are widely suspected to influence political attitudes. However, previous evidence from switching off the algorithm on Meta platforms found no political effects'. Here we present results from a 2023 field experiment on Elon Musk's platform X shedding light on this puzzle. We assigned active US-based users randomly to either an algorithmic or a chronological feed for 7 weeks, measuring political attitudes and online behaviour. Switching from a chronological to an algorithmic feed increased engagement and shifted political opinion towards more conservative positions, particularly regarding policy priorities, perceptions of criminal investigations into Donald Trump and views on the war in Ukraine. In contrast, switching from the algorithmic to the chronological feed had no comparable effects. Neither switching the algorithm on nor switching it off significantly affected affective polarization or self-reported partisanship. To investigate the mechanism, we analysed users' feed content and behaviour. We found that the algorithm promotes conservative content and demotes posts by traditional media. Exposure to algorithmic content leads users to follow conservative political activist accounts, which they continue to follow even after switching off the algorithm, helping explain the asymmetry in effects. These results suggest that initial exposure to X's algorithm has persistent effects on users' current political attitudes and account-following behaviour, even in the absence of a detectable effect on partisanship.

A new paper shows that less than 2 months of exposure to Twitter’s algorithmic feed significantly shifts people’s political views to the right.

Moving from chronological feed to the algorithmic feed also increases engagement.

This is one of the most concerning papers I’ve read in awhile.

Anthropic and the Pentagon are reportedly arguing over Claude usage | TechCrunch The Pentagon is pushing AI companies to allow the U.S. military to use their technology for “all lawful purposes,” but Anthropic is pushing back,

The Pentagon is pushing AI companies to allow the U.S. military to use their technology for “all lawful purposes,” but Anthropic is pushing back,

There’s no such thing as an unbiased chatbot. Academic librarian Raina Bloom @mmelibrarian.bsky.social on how “AI” developers ignore how people actually become informed. From Ep. 51 of Mystery AI Hype Theater 3000, “The War on Knowledge”.

(with @emilymbender.bsky.social)

Apple ‘runs on Anthropic,’ says Mark Gurman - 9to5Mac Mark Gurman is known as being well connected inside Apple, and he just shared some interesting comments about Apple’s reliance on Anthropic.

Apple runs on Anthropic internally, but did a deal with Google/Gemini for its products.

a cartoon cat with big green eyes and a black bow tie ALT: a cartoon cat with big green eyes and a black bow tie

I say do it, but with specific consequences for breaking the rule again.

When my daughter was young, she figured out that if she just did this, I would usually fold anyway.

Woolen….

Thank you for this. I needed a good laugh this morning.

Congress calls for expanded multilateral defense ties across Indo-Pacific The latest U.S. defense policy signals Congress expects deeper cooperation with allies across the Indo-Pacific to counter an increasingly assertive China.

The latest U.S. defense policy signals that Congress expects expanded multilateral defense ties and deeper cooperation with America's friends and allies across the Indo-Pacific to counter an increasingly assertive China, according to defense experts. (from @brianmcelhiney.bsky.social)

this is worth reading!

there’s always been a class of throwaway code, but it typically was only accessible to top engineers. AI opens that up to many more

but does that mean engineers go away? maybe it’s a complex question

AI may not need massive training data after all New research shows that AI doesn’t need endless training data to start acting more like a human brain. When researchers redesigned AI systems to better resemble biological brains, some models produced...

"When researchers redesigned AI systems to better resemble biological brains, some models produced brain-like activity without any training at all."

The statute of limitations for violations of federal law is 5 years.

Your photographs, videos, and livestreams can be used as evidence in future cases.

So while this Administration may not hold anyone accountable, a future Administration will.

So, the answer engine will provide you answers and ads, and presumably over time, when enough money is paid, the answer will be the ad.

Elon Musk says retirement savings 'won't matter' in 20 years. We asked 7 personal finance and AI gurus what they think. — Business Insider Updated Jan 15, 2026, 10:01 AM GMT The Tesla board has proposed a pay package for Elon Musk that could be worth up to $1 trillion. ALLISON ROBBERT/Getty Images; Getty Images; Tyler Le/BI Elon Musk sai...

Complete BS - “The world's richest man predicted that advances in AI, energy, and robotics will generate such an "abundance" of resources for all that individuals' retirement savings will be "irrelevant.”

State of AI Search Optimization 2026 Search is shifting from ranked lists to definitive answers. This guide breaks down the retrieval, citation, and trust factors that determine LLM visibility in 2026.

Answer Engine vs Search Engine

Air Force says AI tools outperform human planners in 'battle management' experiment - Breaking Defense In a complex multi-service, multi-domain scenario — not typically covered in Air Force training — the best of several AI tools generated tactical “courses of action” not only faster than military...

The norm has been/was about 3-5 COAs with at least one being a throwaway - "It found that at least one of the AI algorithms not only generated more COAs in less time than the humans, but it actually made fewer errors than the humans did as well."

In 2026, AI will move from hype to pragmatism | TechCrunch In 2026, here's what you can expect from the AI industry: new architectures, smaller models, world models, reliable agents, physical AI, and products designed for real-world use.

AI will move from hype to pragmatism | TechCrunch

Pentagon task force to deploy AI-powered UAS systems to capture drones The DroneHunter 4700 is a lightweight system that uses AI and radar to identify, pursue and capture other drones using an array of expanding nets.

DroneHunter is a system that uses AI and radar to identify, pursue and capture other drones through an array of expanding nets.

What Are the Top Defense Technology Priorities for 2026? A Quick Guide In the latest article from IDGA we take a look at the technologies the DoW is likely to prioritize in 2026.

So this lists AI, Hypersonic weapons, Supply Chain, Counter UAS, and some Navy and AF air platforms.

Followed

Edited it: “the Smithsonian *temporarily removed mention of Trumps two impeachments….”

EU Regulations Are Not Ready for Multi-Agent AI Incidents | TechPolicy.Press With the AI Act's Article 73 guidelines set to become binding in August, the clock is ticking for the European Commission to address loopholes.

Incident reporting under the EU's AI Act will become binding in August, but the draft guidance still misses multi-agent and cascading AI failures, write Natàlia Fernández Ashman, Usman Anwar, and Dr. Marta Bieńkiewicz, urging the European Commission to close the accountability gap.

Grok is an Epistemic Weapon | TechPolicy.Press Elon Musk's Grok claims to be truth-seeking while shaping discourse and amplifying ideology, argues Matthew Kirschenbaum.

Grok stands as the clearest example of an AI that is both ideologically aligned and operationally embedded in a social media platform, says Matthew Kirschenbaum. Its mechanisms should be taken apart under bright lights, in much the same way people learn how to defuse a bomb, he argues.

Microsoft warns Chinese companies, especially DeepSeek, are winning AI adoption outside the West, gaining significant market share in the Global South (Cristina Criddle/Financial Times)

Main Link | Techmeme Permalink

Australia social media ban: Meta blocks 550,000 accounts under new law The new law, which came into effect in December, is aimed at protecting children from online harm.

"About 550,000 accounts were blocked by Meta during the first days of Australia's landmark social media ban for kids." www.bbc.com/news/article...