I'm a fan of this @binarybits.bsky.social piece that points out the incredibly simple and dumb error made by shadowstats. There's definitely some wiggle room for how you calculate inflation but you can't be significantly biased in the long term without it being obvious.

Great @binarybits.bsky.social piece with overview of and background of White House AI Action Plan www.understandingai.org/p/donald-tru...

The American Time Use Survey came out recently, giving the most detailed look at how the country spends its days, so here's a quick thread 1st, the share of Americans reading for personal interest on an average day has fallen to the lowest level on record, down 10% from 2003🧵

This is an excellent explainer by @binarybits.bsky.social . arstechnica.com/ai/2025/07/h...

The Truth about Sam Altman | @robertwrighter.bsky.social and Timothy B. Lee (@binarybits.bsky.social), author of the Understanding AI newsletter, discuss what two new biographies reveal about the OpenAI CEO.

Fantastic @binarybits.bsky.social writeup of the Alsup ruling in Anthropic’s case open.substack.com/pub/understa... But aside from AI training, yet another reminder of the power of analog media. This excerpt is tragic: Alsup’s right about books. But this is exactly how the digital economy works.

No urbanist has successfully kept more vehicles off the road than radical degrowth war-on-cars advocate Donald J Trump

Apropos of nothing I feel like one of the most important things to help your kids develop is an ability to self-regulate.

Hey @niedermeyer.online is this going to mean no more Tesla takedown protests? Or are they going to become pro-Elon rallies?

Groceries stores are famous for their fat profit margins.

Great article on Waymo safety by @binarybits.bsky.social, who has IMO the best reporting on self-driving cars out there

Interesting look at the emergency safety regulations that may hamper rollout of the promised Dallas to Houston automated trucking trial from @binarybits.bsky.social.

My brother is hiring!

@binarybits.bsky.social is doubtful that scaling up so-called reasoning AI models will "achieve human-level intelligence" given this "probably has more than two dimensions, and mastering all of them will probably require a few more conceptual breakthroughs." open.substack.com/pub/understa...

19 professionals (in a variety of fields) evaluated OpenAI's Deep Research vs Google's Deep Research. OpenAI was the clear winner 🏆 Neat study by @binarybits.bsky.social, read more here: www.understandingai.org/p/these-expe...

I was a guest on the AI Summer podcast the other day. I go into some detail about how the Shortwave AI agent works today, how we got here, and what's next. Check it out: www.aisummer.org/p/andrew-lee... Thanks @binarybits.bsky.social and Dean Ball for hosting me!

The Understanding AI newsletter from @binarybits.bsky.social really helps you to, well, understand AI. You should read it. www.understandingai.org/p/these-expe...

This is what Elon does: he takes something someone else is already doing, does it more sloppily/dangerously, and slaps his own brand on it so everyone forgets anything else exists (if they ever knew). You can't cede his rebranding without a fight, or you're toast. Explain that GAO already does this!

Want to keep up with the latest developments in AI? Check out the 7 newsletters I personally read & recommend for AI news, insights, and analysis: 👉 www.dataschool.io/best-ai-news... 👈 Includes: @simonwillison.net @emollick.bsky.social @sebastianraschka.com @binarybits.bsky.social

Interesting theory and excellent DeepSeek explainer by @binarybits.bsky.social www.understandingai.org/p/i-dont-bel...

Excited to launch a new podcast with @deanwb.bsky.social about AI. It's called AI summer, and can be found on all major podcast apps. The first episode is an interview with political scientist @jonaskonas.bsky.social about AI policy in the Trump era. www.aisummer.org/p/jon-askona...

If TiKTok wasn’t a benefit for Chinese intel, they would sell. They could make a killing. But it’s not about the money

Nobody knew the federal budget could be so complicated.

Here are some interesting opportunities to collaborate with the excellent @binarybits.bsky.social on his newsletter, Understanding AI, if you or someone you know are interested in serious AI journalism. If not, I still recommend his newsletter. www.understandingai.org/p/two-opport...

I’m not sure what industry “four trending stories daily” is the standard for, but it’s definitely not journalism.

Donating to the Wikimedia Foundation is great. Becoming a volunteer editor is even better. youtu.be/bRRHR1NEOqE

She’s mangling history here. Truman was actually Secretary of State (he would later become vice president and then president upon the assassination of James Garfield in 1881). The idea was so unpopular (it was called “Truman’s folly”) that president Monroe had to drop the idea.

Tfw the book is so tedious you bump it up to 2.5x.

Elections have consequences.

Just retired & looking back writing code is the least of it. Good abstractions/design, mentoring, writing *useful* docs, debugging large distributed multi-threaded code in different containers/platforms, negotiating with both tech and non-tech teams, customers, researching new things took most time.

I have updated my disclosure statement. timothyblee.com/disclosure-s...

This is a funny joke but it also reflects biased media coverage. Conventional cars catch on fire all the time. They just don’t get covered because it’s not considered news. When i was at ars I passed on most Tesla battery fire stories for this reason.

Official data released today shows US data center construction hitting a new all-time high of $31.5B annualized, up 43% from this time last year and 128% since the release of ChatGPT two years ago

This Texas bill sounds worse than SB 1047. Kind of incredible that it was introduced by a Republican. www.hyperdimensional.co/p/texas-plow...

I was following the fake one!

What’s the deal with mastodon these days? Has it been losing ground to bluesky? Does it have a different position on the ideological spectrum? It seems likely that one of them will wind up as the default left-of-center Twitter alternative.

I just don't see H sapiens ceding decision-making, directive authority to machines. "But what if the machines make better decisions? Competition will force ..." Look around this ridiculous planet and ask yourself how often competition has forced hominids to cede power to the best problem-solvers.

It’s an interesting marketing choice to make all your big product announcements in the two weeks before Christmas.

Mary and Joseph had to stay in a manger because Bethlehem resorted to anti-Airbnb rules to boost housing affordability rather than rolling back Herod’s anti-supply regulations.

Most businesspeople are always going to prefer Republicans because of low taxes, but the post-2012 effort to *deliberately stigmatize center-left business people* was a disaster. www.slowboring.com/p/liberalism...

Some people are absolutely doing this for careerist reasons but also think it’s healthy for supporters of a losing candidate to honestly examine what they might have gotten wrong. The voters aren’t infallible but if a bunch of former supporters turn against you there might be some good reasons.

The key to low-crime cities is more police, not more incarceration. This might be the area where wokeness has most distorted Democratic policymaking. I would love to see a Democrat in 2028 run on bringing back Bill Clinton's more-cops-on-the-beat agenda. www.noahpinion.blog/p/good-citie...

If you're not obsessively following AI, strong recommend below for implications of the latest models focusing on a new type of training and compute paradigm. Clearly explains what's different and the implications. Worth the subscription just for this article IMO. open.substack.com/pub/understa...

Here’s a fun and (I think) generally accurate taxonomy of AI skepticism. buildcognitiveresonance.substack.com/p/who-and-wh...

@robseamans.bsky.social you’ve got a not-very-subtle impersonator.

Can someone explain the threat model behind this warning? It has never occurred to me that loose fitting clothing would be a fire risk. Is the idea like the garment catches fire when the child reaches over a candle?

I don’t understand why people think drones in the sky are a mystery to be solved. Millions of people own them. It is not surprising that when they get used people nearby notice and don’t know who is using them or why.

Expects making mistakes isn't new. What's new is that the Internet has made it far easier for non-experts to publicize and complain about their mistakes. Which makes it seem like today's experts are worse than those of the past. www.slowboring.com/p/have-exper...

I am not worried about AGI or superintelligence. We already live among superhuman machines. They are stronger and faster than me. They can FLY. There are still plenty of people around When AI can perform superhuman intellectual labor, we will use it to build amazing things that are impossible now