Item #1: “A positive-sum outlook on the World.” I usually agree with him but disagree here:
All the best people have an incredibly positive-sum outlook on the world.
They help other people to achieve more, do better, find love, and so on.
And their lives are enriched by it.
— Crémieux (@cremieuxrecueil) December 6, 2025
This is good advice in theory, but what can also explain how people who are also so persistently negative can be really successful too, beyond survivorship bias. I see this a lot. Negativity seems to dominate so much online, and those people who instigate are successful. The belief that these negative people get their ‘just deserts’ is wishful thinking. Andrew Tate proved you can be extremely successful roleplaying as the villain.
This disconnect is also an example of revealed vs stated preferences. People state they like positivity, but their actions or content consumption habits suggest otherwise or it’s unclear. The fact it went viral is also illustrative of the social desirability bias, in that ‘sharing and liking it’ projects positive traits; a form of ‘virtue signaling’, if you will. It’s like when you say “my condolences” at the news of someone dying. It feels good to say but doesn’t do anything.
Item #2: AI vs math. There has been a lot of hype on Twitter about AI making math breakthroughs and solving some open problems. A lot of problems remain open because no one has actually worked on them. A conjecture by definition is an open problem. My take is, AI is a long way from obsoleting mathematicians. I can’t put a specific timeline on it, but in 2-3 years I predict progress at AI-assisted math will have stalled or failed to meet lofty expectations.
I’m not gonna lie: I am occasionally blown away by AI. I use Chat GPT for some of my own problems, such as solving systems of equations where Wolfram Alpha falls short. Chat GPT v.5 represents a marked improvement over v.4. For example, v.5 is able to solve a certain definite integral exactly instead of only approximating it, whereas v.4 failed to give the exact answer.
But actually solving or proving something abstract–with many non-obvious steps–is a whole other ball of wax. For the math paper I finished, it involved deriving a unique expression which has hundreds of characters and required specific, non-obvious substitutions. Math is very subtle at the research level. To say it’s like finding a needle in a haystack undersells the difficulty, because there you at least have some bound. It’s exceedingly improbable an AI will recreate my result. I fed it an elementary version of my problem and it proved it in a minute, which is still pretty impressive that it did it at all.
Item #3: Bitcoin vs QQQ.
This is literally fake news from slate.com: “Trump Gave the Tech Bros Everything. Why Are They Still Crashing Out?”
Crashing out? Tech stocks as measured by QQQ are up huge this year. There is no AI bubble except the one invented by the media. The only thing that is crashing is crypto, exactly as I predicted. See below the 1-year performance of Bitcoin vs QQQ and how they diverged:
Again, I was the only one to predict a crypto crash and a tech stock surge, and no AI bubble. Everyone else either predicted the stock market would crash or bitcoin would surge, or some combination therein. Trump is fully on-board tech, but has kicked crypto to the curb. There has been no BTC reserve and the stockpile remains unfunded. Basically, no progress after a year despite crypto donors rolling out the red carpet for Trump in 2024. The liberal media is so desperate for an AI crash. Their jobs are threatened by AI, so they are inventing fake news to try to make it a self-fulfilling prophecy.
Item #4: Interesting video by Nick Norwitz: “Calories Don’t Cause Obesity…Yes, Really”
There is a persistent myth that when you overeat at once, like binging, that your body does not use all the calories. “Use” in this context is somewhat confusing. Generally speaking, 90-95% of calories you ingest are digested, but how the body uses this nutrition when in a surplus is highly variable. For someone with good genetics for leanness whose body naturally resists weight gain, the surplus is converted to ‘waste heat’ instead of fat. This is apparently the case for Nick Norwitz, who is able to eat a lot and still stay lean. Most are not this fortunate.
He mentions undigested fat in the excrement. But, no, surplus calories don’t get ‘shat out’. If you were shitting undigested fat it would be very obvious, as consumers of those infamous Olestra chips during the ’90s could attest. It floats like an oil slick and is impossible to flush. Again, 90-95% of ingested calories are digested, and can be stored as fat. The body does not simply decide to ‘refuse to digest food’. If this was to happen it would lead to bloating, diarrhea, vomiting and other obvious and unpleasant symptoms. Acute overeating can cause vomiting, but this isn’t the same as the body simply refusing to digest the food.
So what happens is as you binge, some of the food is stored in the stomach, and the rest is quickly shunted through the small intestine. This is an enormous organ that has huge surface area and can expand considerably to accommodate as much food that is required of it, which is eventually all digested, with some percentage of calorie surplus showing up as fat days later. The overheating Nick feels is probably related to his body generating waste heat.
Item #5: Weight loss challenge update: Fail (so far)
With the IQ and trading challenges complete, I embarked on the weight loss challenge last week. Progress has stalled because I was unable to stick to the plan and overate, as I predicted last week. I am only giving myself a 10-30% chance of success. I hereby restart my plan. This will be the hardest mentally of the three challenges, and I am trying to set realistic expectations here.