I could do my job in my sleep. Oh wait, maybe I am doing my job in my sleep.

A strong neck helps if you drive a car at a speed of 200 mph. Adults with “tech neck” don’t work in such extreme conditions, but experts say that adapting the training techniques of professional race car drivers might alleviate the side effects of prolonged computer, tablet and smartphone use.

Fantastic (long) piece on what AU means to mathematics (and what mathematics is) So many insights. The first begins why so may people in AI research are getting obsessed with “taste”. davidbessis.substack.com/p/the-fall-o...

After grumbling for years in our substacks about uses and abuses of probability in machine learning, @beenwrekt.bsky.social and I decided it's finally time to put down some concrete ideas in a manifesto with a DOI. Here it is, give us your best feedback. (1/4)

To accompany my textbook (Computational Foundations of Cognitive Neuroscience) and the class I taught this semester, I'm open-sourcing my lectures slides: gershmanlab.com/lectures.html I'll continue to update these as I improve them.

Cleaning up yet another agentic ai mess recently, I was struck by an awful thought - *it’s Lotus Notes all over again*.

"This is not making America great again, this is evil" Like @markhamillofficial.bsky.social , I am stunned to find myself completely agreeing with Marjorie Taylor Greene. She is right, though, so here I am.

arstechnica.com/tech-policy/... Sounds like someone else we know….

Does not look too good for the home team here in AZ. We hit 100 in March this year. Not looking forward to the summer. Pattern is interesting across states though. share.google/HAYXjszdxox5...

Related read: ergosphere.blog/posts/the-ma...

arstechnica.com/ai/2026/04/r... If you s/LLM/consultant uniformly here you see this is not new. And the tremendous success of both depends on the high propensity of executives to offload.

Given the function specifying a sigmoid-MLP f(x)=Wₙσ(⋯σ(W₁x+b₁)⋯)+bₙ, we can recover the weights/biases up to (unique) signed permutations of the neurons by looking at f over the complex numbers, see the quoted thread. What happens if f is odd, i.e. f(x)=−f(−x)?

It’s easy to ignore the hedgehogs - until they bloom!

Real tech companies learned 20 years ago that OSS always wins.

Really disappointed by the Warren endorsement of Platner. Get that it's a (needed) challenge to Schumer, but it doubles down on an anti-fascist strategy that has failed when you have the Minnesota model *right there* Older, longer thoughts here: cooperativeoverlapping.substack.com/p/on-wine-mo...

My brother once said that his HS education was no better than sitting inside with the shades drawn watching “I love Lucy” reruns. That’s more or less how we train LLMs. Good thing ppl like @nsaphra.bsky.social are getting serious about LLM training dynamics. youtu.be/2WqvUQY1nlg?...

The cryptocurrency industry super PACs dumped $14.2 million into the Illinois primaries. 90% of that – $12.8 million – was wasted, in that it went to opposing Democratic candidates who won their primaries (Stratton in the Senate race, Ford in H-07) or supporting their opponents.

This is painful to read. Every gifted woman who has left mathematics has her own story. They are not all like this one. They are not all the same. But the result is always the same: they take their gifts elsewhere.

This looks actually very similar to coding - big risk of massive quality decline and unenlightening dead ends, but also potential to massively improve productivity of human mathematicians. Comes down to who is driving and staying in control.

Marc Andreessen's career success has mostly been a result of fortuitous timing and circumstances. Very little actual skill involved. So now his whole worldview orbits around *ignoring all of that.* ...Why dwell on the minefield of all the dumb-and-wrong choices he made? Bobble-headed chucklefuck.

www.alexerhardt.com/en/enshittif... The natural conclusion.

“What is cross-entropy loss and why do we call it that?” is my favorite weed-out question for AI/ML jobs. Fail cdf looks like this How to compute in python: .75 Definition: .9 Why good for classifiers: .95 Real understanding incl why “entropy”: .99 Depressing but effective.

This is telling coming from a leading LLM researcher.

fuck this so much we are so not prepared for the mountain of dogshit “scholarship” about to flood journals everywhere

1/ I am constantly droning on about how AI ppl need to understand what their models are doing. I am starting to lose the battle when it comes to model selection - “I can just ask Claude and it will select the right model for me, so stfu about me needing to understand what it is doing.”

Let me ask the question that is going to have me burn on the pyre: why do we need models to learn good representations already? Why does it matter?

My take on possible oil scenarios & conflict in Middle East. Inspired by Jurassic Park and Jeff Goldblum’s explanation of chaos theory. This one hit close to home as I know too well tail risks. My mom cried - should have warned her first. Lots of humility amidst uncertainty.

I really hope that this does not turn out to be a Claude targeting mistake. We will likely never know, but this is precisely the reason that Dario did not want Anthropic tech used without humans in the loop.

Wow

This looks interesting. I have always hated vector databases because they force stock embeddings. github.com/GoogleCloudP...

I thought about taking the flag down today out of shame. But then I worried about what my neighbors would think - not in the way one normally uses that phrase. I was afraid that they would think that I had given up. I am not going to give up on my neighbors.

uphack.io/blog/post/th... This is excellent.

www.tandfonline.com/doi/full/10.... Saw this in dead trees version which is ancient by now, but the approach is cute. Claude Sonnet 3.5 smokes same epoch other models on 0-shot slop classification. Error rates are not impressive though.

otava.apache.org Great to see leading performance engineering tools coming to the ASF

After clearing 2 rodent nests there, we were actually happy to find “security” under the AC unit yesterday. Loud rattle is a plus.

When I get bored with science, it's because I've read the same paper three times from different authors whose media diet consists of corporate PR releases and unhinged slop summaries for viral arxiv drops.

It’s hard to image any leadership job in any company where Trump’s no plan, grab-ass leadership style would not get you fired. This is so embarrassing.

Go work with Ted. You will learn something.

Integrating AI-driven risk detection with scheduling systems could enable construction projects to automatically adjust plans in response to emerging issues, potentially reducing delays and improving resilience.

I am tempted to try the problem :)

Beautiful day for a run

www.quantamagazine.org/the-man-who-... This article really bugs me. Missed opportunity to explain the arguments. Smear piece distracting from what must have been a great collaboration. I expect more from Quanta.

gist.github.com/dollspace-ga... We can actually do this kind of thing now. Let’s harvest the AI surplus to improve software quality.

Any sufficiently large k-nn is indistinguishable from magic 🧙‍♂️

When a company in an industry built on hype tells you that a use case is a bad idea—and actually dangerous—that means it’s a *catastrophically* bad idea.

While whiskey Pete plays army with Grokk and shoots at speedboats, our defenses against today’s actual imminent threats fall into disrepair.

Now THAT's a headline. "The U.S. spent $30 billion to ditch textbooks for laptops and tablets: The result is the first generation less cognitively capable than their parents" fortune.com/2026/02/21/l...

Did ChatGPT write this talking point for you, Sam? Or do you just *organically* suck this bad?