Dean Eckles

We're excited to be hosting CODE@MIT again this fall! We invite a mix of work involving experimentation — methods, cool experiments, industry reports, etc. Look for more info later on this year's invited speakers.

The only reason anyone is talking about Marissa Mayer or Dazzle is that her marketing team understands how to deploy rage bait, and most of the Internet still can't spot it in the wild.

Excited to share that my paper with Zhongren Chen on generalizing causal effects in the presence of non-compliance is now out in Political Analysis! doi.org/10.1017/pan....

“But surely all academics want their papers to be read” you cry. Some certainly do. But a lot of people actively dislike it. They want them to be counted, of course, in various ways—n citations, impact of outlet, etc—but you might be surprised at how many would prefer people not actually read them.

I get 5+ emails *a day* from prospective PhD students. I have seen some advice going around about what to write in these emails. Here's my piece of advice: don't waste your time. Or your tokens. #phdadmissions Or, if you really insist...

Side benefit of this idea: I can direct prospectives to less established faculty they might not be aware of

"No matter how we look at the data, AI wins. It wins in 100% of the pairs overall when judged by another AI model...and 87% overall when judged by human journal editors... [H]umans agreed as much or more with AI’s judgment than they did with other humans" gking.harvard.edu/files/peerle...

a properly aligned ai would never agree to make a pie chart

It's not even A GOOD GRAPH

The (V,E) people got the name first. Sylvester borrowed "graph" from chemical bond diagrams. Plotted functions got called graphs shortly after. mathshistory.st-andrews.ac.uk/Miller/mathw...

Even more to the point, directly labeling persuasive text as Ai generated doesn't change its persuasiveness academic.oup.com/pnasnexus/ar...

Some notable examples of long-standing errors in science from Douglas Allchin’s new article: “When is science ‘self-correcting’?”: doi.org/10.1016/j.sh... Few quotes follow…

We are very nearly profitable, we just need a one-time capital outlay to build a virology lab so that we can tell whether or not Claude is aligned based in whether he chooses to use it to create covid-26 or not

I’m really incurably looking for potential silver linings and i think it is possible that AI’s overly smooth blandness helps more people understand the point of experimental and chaotic prose forms

the routine, pay for play pardons are all by themselves one of the worst corruption scandals in american history!

Today we launch the first episode of my podcast. I interviewed Whitney Mitchell, whose fiancé and caretaker Garrett Foster was murdered at a BLM protest in 2020. Foster's killer was convicted by a jury and given a 25-year sentence. Then Texas Gov. Greg Abbott pardoned him.

First report from my sabbatical year, during which I am traveling to 6-7 universities around the globe to teach a three-day course on AI & Society. Where better to start than Bhutan, the remote Buddhist kingdom in the Himalayas? I wrote it up here (part 1). www.linkedin.com/pulse/ai-rin...

Why are private equity execs flocking to Question 3? It's because they think it's a good for their bank accounts and a way to crush nascent working-class populism. I unpack it in the piece here. working-mass.com/2026/10/19/o...

Amazing. Also, 42 lmao

Did you know that they actually had a resuscitator so that the bird would survive??? blog.scienceandindustrymuseum.org.uk/canary-resus...

Did Ukrainian drone strikes really silence Russia's social media bots? (No) A real Cloudflare chart fueled a viral story ab MAGA and AfD accounts going quiet. But the evidence doesn't support the rumor. I wrote about bots, vibes, & what we've lost as platforms have shut down tools & data access.

Fascinating if real, but I suspect most people have no idea if they live within 2 km of a data center and this represents factors that are correlated with proximity

“Yum. Give me that Hieronymus Bosch burger”

These different incentives from standalone media owners vs. "common owners" like Benioff is the subject of one of my working papers www.dropbox.com/scl/fi/07aux...

Getting Instagram ads for my failed experiments now

This thing has made me go do a lot of reading tonight, and the Wikipedia page about the last public execution in the United States is well-written. It was Rainey Bethea, a Black man hanged in front of a crowd of 20,000 in 1936, a spectacle which embarrassed Kentucky enough that they ended hangings.

No artificial sweetener for me, thanks. I'll have the super sweetener.

In an authoritarian regime, people publicly debase themselves to signal loyalty to the ruler

Something not sufficiently registered in the convo around AI/IP: the value that was gained from pretraining (on our collective legacy) is now common property. People do pay more for frontier models than open ones — but the differences they're paying for, at this point, come from post-training.

Models like Talkie-1930 sound like voices from the past. If they could reliably speak from specified historical vantage points, researchers might also use them to simulate the past. But how reliable are they? Today we release a benchmark answering that question for English contexts 1831-1930.

Trump’s insider trading is, on its own, the worst corruption scandal in US history, with $$$ dwarfing Teapot Dome. So are Trump’s bribes-for-pardons. So is the Qatar bribe plane. So is all the Saudi money he (and Kushner) have pocketed. And the ballroom. And more. Each alone would be the worst.

“That’s quite an act, what do you call it?” “Pacing the frontier!”

A very important message from Omer Reingold to our TCS community, especially us (arg, already) senior researchers: "snap out of it." Please, digest, and share. theorydish.blog/2026/09/29/s...

Part II in @kkomaitis.bsky.social epic 10 part series about how and where big tech went wrong. I promise you, you're going to want to read the whole thing...

It's the end of an era. The Social Media Collective is shutting down. Thanks to everyone who was a part of this crazy endeavor, but especially @nancybaym.bsky.social, Tarleton Gillespie, and @marylgray.bsky.social. I adore you all! socialmediacollective.org/2026/09/23/s...

"Why is There so Little Money in U.S. Politics?"

Here's a fun question - why does the broad US public have an unusually *high* belief that "social security and pensions" will result in an adequate retirement right now? data.sca.isr.umich.edu/get-chart.ph...

I'm hiring an economics predoc at @columbiasipa.bsky.social, generally on applied micro projects, but with a large focus on the impacts of AI on economic outcomes using large-scale field experiments with many companies. Please share widely: apply.interfolio.com/194121

Interesting: Perhaps you know that Thomas Carlyle coined "the dismal science" for Economics in a speech fighting against slave emancipation. I didn't realize he was also anti technology: "Our true Deity is Mechanism. It has subdued external Nature for us, and we think it will do all other things."

Trump misstating and touting a meaningless deal over Greenland, while the entire Middle East security architecture falls apart due to his own actions, is emblematic of this administration’s foreign policy.

DHS is actually accusing hundreds of cruise line employees of possessing CSAM materials and revoking their visas without filing criminal charges. It's a way of bumping up their numbers without having to prove anything I wrote about it last year and NPR did this June www.npr.org/2026/06/06/n...

Republicans, in their infinite wisdom, left a big gambling loophole in the Medicaid work requirements. I'm no fan of the gambling industry, but if you are on Medicaid and struggling to deal with the work requirements, taking a $580 bet on both sides of an outcome is one way of getting around them.

NEW: We spent months investigating a little-known media startup called Brown Brothers Media, which buys legit news sites and turns them into AI-powered zombie content farms What blew my mind was its traffic -- it appears to be getting more readers than NPR or CBS News futurism.com/artificial-i...

As models get close to — or, in some cases, beyond — human experts, it becomes very challenging to figure out how to measure them. But what a fascinating problem! I suspect we're going to have to evaluate models the same way we evaluate human innovation: dialogically and perspectivally.