I still remember when Model Context Protocol (MCP) was a big thing. Now it's a relic of bygone days. Things are moving so fast.
Great lesson in data analytics. Neuroscientists using machine learning to sift through a dataset thought they found a biomarker for IQ. It turned out they were tracking socioeconomic status. But even this could be murky. Many people who are wealthier do have a higher IQ, so it may not be an independent covariate. I wouldn't toss the analysis out just because of this. You can give poor people a lot of money and a comfortable place to live, and they could trash it and be dirt poor again by next year. I've found that you need to change your mindset to become wealthy. Teach yourself how to acquire cashflow-paying investments instead of just putting money in the bank. That's the secret behind Kiyosaki's book Rich Dad Poor Dad, and it appears this guy learned the secret, too. So a higher socioeconomic status might be a surrogate marker for mental flexibility and resilience, rather than IQ per se, but having a higher IQ helps.
––– 凄い –––
Why Do We Need Human Mathematicians Anymore? This is actually a column by Po-Shen Loh, the coach for the U.S. IMO team. Boy, I never thought I'd hear the day when mathematicians would say that math progress needs to slow down. How crazy is that?
––– 凄い –––
Great video by Scott Jensen on the stagnation of user interfaces for desktop computing. I agree, when the iPhone and iPad came out, Apple neglected the computer, and focused on the mobile device. I was disappointed. Now that there is money in AI/ML, Apple is now developing the computer again. But innovation on the UI/UX is dead. There is no innovation now. I like Scott's desktop prototype and wish it were available.
––– 凄い –––
ChatGPT now knows what you do on other websites via ad collector. Boy, as if I didn't have enough reasons to avoid using ChatGPT at the public website. So many people want to cross over to the dark side and implement tracking. I use automatic cookie deleters, but who knows who well they really work.
––– 良くない–––
Trump is creating an AI Force, and is against an AI slowdown. There does not need to be a regulatory body ready to shut down bad AI, because you know this can happen. But I do agree that there is no need for the government to shut everyone down right now. That's regulatory capture. If the heads of AI companies want to slow things down, they are in control, and can slow things down themselves.
––– 凄い –––
Your Morning Coffee Might Be Protecting Your Brain from Dementia. I believe we already knew this. Well at least it's nice to be reminded that it still is true.
Speaking of cognitive disability, I saw this page advocating that people use a typeface called Lexend, that would "change the way the world reads" in terms of reading disability. So I was ready to give it a try, but at the bottom, I have to give away my name and email address first. Nope. I don't need to try it this badly. Gee, what a scam.
––– 凄い –––
MIT Technology Review answers your questions about whether AI will kill you. All your questions answered, they say. The folks at Anthropic's METR are working on it. Somehow, I'm not so reassured.
––– 良くない–––
Wow, this will make decisions easier for hematologists and ED physicians: Jehovah’s Witnesses allow blood-derived products but keep ban on whole-blood transfusions. So you can transfuse them now – just not with whole blood transfusions, which is OK, since they are rarely ever done.
––– 凄い –––
In the past 12 months, business applications rose in 47 states. Only 3 states saw less: Washington, Oregon, Hawaii.
I laughed when I read the comment from a Redditor who said that he/she believed that in states where the economy was poor, you would see more business applications as people left to form their own business. 🤣 Yeah, the stuff people believe. When you're in the company of the left coast states, you're probably on the wrong side of sound economic policy.
Oregon data centers by the numbers: 111 facilities consume nearly one-quarter of state’s power, according to new report. A quarter? That seems high. Does that mean Oregon is hardly doing anything else? We'd better get off solar and wind and start generating some serious power. How can we realistically accommodate business growth? Let's not become Cuba.
––– 良くない–––
The Millennium Problems for Biology. it's not from the Clay Institute. I like some of the challenges posted. These are a good start.
––– 凄い –––