Daily Bytes: The Next Batch
Yes, Claude 4.0 can replace Substack's top philosopher
I’ve been procrastinating and desperately trying to emerge from a rabbit hole I crawled into several weeks ago when I started pondering issues arising from the maddening “Mary’s Room” thought experiment. I continue to read and occasionally comment on other people’s posts though, and somehow got fed a Note by Dan Koe advising me how to listen to my inner voice and find my true path (and make a lot of money in the process, as he is doing.) I read several more of Dan’s posts (here’s one: A helpful note from @theDanKoe), and I thought, “I’ll bet an AI could read a handful of these and then churn out dozens more.” Challenge for my sometimes friend Claude 4.0. I uploaded 25 examples of Dan’s Notes, asked Claude to characterize them, and then requested that he take over the account for 5 days, so that Dan could take a break without disappointing his followers. I think Claude nailed it! Here are the 5 Notes it came up with. (Do read a few of Dan’s notes if you’re not familiar with them so you can compare.) I don’t present these to demonstate that LLMs are clever. There’s a rather different conclusion one can arrive at, but I’ll leave that to my dear readers.
Back soon with a real post!
For your consideration. The next 5 aphorisms as presented by a stochastic parrot named Claude:
Monday: Most people think they need permission to start. They wait for the right moment, the perfect plan, the blessing of someone important. Meanwhile, the people changing the world realized years ago that permission is something you grant yourself. The market doesn't care about your credentials—it cares about whether you solved a problem that matters.
Tuesday: You know that voice in your head that whispers "this is impossible" every time you think big? That's not wisdom. That's your brain trying to keep you safe in a world that no longer requires safety. Your ancestors needed that voice to avoid getting eaten by lions. You need to ignore it to avoid getting eaten by mediocrity.
Wednesday: The reason you procrastinate isn't because you're lazy. It's because deep down you know the thing you're avoiding isn't actually important to you. Your subconscious is smarter than your conscious mind—it refuses to waste energy on things that don't align with who you're becoming. Stop fighting the procrastination. Start questioning the goal.
Thursday: Everyone talks about work-life balance like they're two separate things. That's the problem. When your work becomes an extension of who you are—not something you do to pay bills—the concept of balance disappears. You don't need to balance breathing with living. You don't need to balance creating with existing. Find work that feels like play, and you'll never work another day in your life.
Friday: The internet rewards people who think out loud. Your random 3am thoughts about why restaurants are designed wrong or how dating apps could be fixed—those aren't just thoughts. They're the beginning of every business that will matter in 10 years. The difference between dreamers and builders is that builders tweet the dream, then spend the next 1000 days making it real.
It’s really uncanny how close these are to the nuggets that Dan churns out for his 131,000 followers. QED (and again, I’m not touting LLMs)
Well if Dan were to let Claude write some of his posts, I doubt his followers would appreciate not being told. Letting AI do cover art is one thing (hopefully attributed), or even non acknowledged research to help write a given post. But if he were to pass off AI statements for his own, that would be full plagiarism. Mind you that I do now find some of my discussions with AI to be at least as productive as my discussions with people. But people are conscious and thus value driven entities to potentially sympathize with. AI are simply algorithms and so feel nothing. Conflating them should be considered a category error.