Chris

Most people are embarrassed by what they haven't done yet. The career they've half-built. The side income they've talked about for 3 years. The version of themselves they keep postponing. I help people close that gap. If that's you, DM me, or book using the link below:

The search function on Substack needs improving.

"In 1850, it was thought among British elites that obviously we should maintain the two-power standard for the Royal Navy. In 1950, this idea was dead. In 1750, Europe elites saw written constitutions as an idea of the devil. In 1850, European elites felt that written constitutions could hardly be resisted. In the mid-19th century, European elites had come to see free trade as mutually beneficial and the future. By 1900, protection had spread and free trade ideas were under attack everywhere." If you're resistant to change - you're resistant to life.s

"You’re better off having the paid versions of Opus or GPT work for you than ~99% of MPs. Question to ponder as reading: how many MPs’ working 12 hours per day for you would be roughly as valuable in political research as paid-Opus or GPT working for you, with token limits set to the cost of the MP salaries pro rata?" Taken from Dominic Cummings' Substack

I lost £167 last night, but it was a useful experiment. I used AI as a predictive model to bet on football fixtures. Football is a sport I already know a lot about (I played it and watched it religiously growing up). I asked AI to place a bet of £5 on Thursday evening's football matches, predicting the most likely way I'd receive £500 in winnings. I placed the bet, but cashed out (Braga scored in the last minute). If I didn't cash out, I'd have won circa £170. Moral of the story: This is just one attempt, but AI's predictions were remarkably accurate. It admitted this wouldn't have won me £500 - but it would've won me circa £170 if I had held my nerve. Is gambling a mugs game anymore?

Dominic Cummings on AI models, and a useful experiment: - Start using AI for topics you already know a lot about. This will help you sense check the accuracy of the output. - Use AI to steel man arguments. This will be particularly useful in any advisory profession like law/consultancy. Sometimes in these fields, there are multiple right answers. Using AI to steel man opposing arguments has huge potential for providing better advice. Paradoxically, I think this is a real threat to the profession. Much of what happens inside law firms (and between them) is just pointless gesturing and arguing based on a deep desire for self-importance, rather than the truth. - You can use this for your life, too. Get AI to ruthlessly critique your life now, and strategise for improvement. This obviously shouldn't be therapeutic, just purely

@krishnanrohit It's important to use AI to steel man opposing hypotheses. AI can be a real enhancer for critical thinking but only if you use it properly.

Something I was wrong about with AI psychosis, a badly named but real thing, is that I thought smart or educated people would easily see through the machinations so won't be fooled. But the sociological miss was that if you're in a position of power, quite often you're used to asking people to do things for you. And assessing their outputs. But this worked because you rely on those people to push back on your worst ideas or water them down. With codex or claude code, that doesn't really happen. You ask them to prove the new theory of aether they jump to it with gusto. They're wonderfully compliant. This means you have to act as your own critic, which many folks in the senior echelons have never had to do. Especially in tech, where being a visionary was of outsized value. The outsourcing of criticism had enormous advantages in that it allows you to think big and come up with crazy visions but until the models push back on them, strongly and regularly, they look at their handiwork and go "look, I knew it all along, I'm a genius".

Everything you're stressed about right now has happened before. To bigger people. With more to lose. They're gone. But you're still here. Zoom out. Gain perspective. Then decide.

Bismarck on politics, but applicable to life: "This trade teaches that one can be as shrewd as the shrewdest in this world and still at any moment go like a child into the dark."

"Politics is a thankless job because everything depends on chance and conjecture. One has to reckon with a series of probabilities and improbabilities and base one’s plans upon this reckoning… As long as he lives the statesman is always unprepared." - Bismarck When life is ultimately politics - this is important to know. A sense of preparedness in life is rare. We have to act without knowing all of the facts.

"A great art you see in very rare people brings together taste and long-term planning, but the word ‘planning’ is tricky because there are a few things that don’t/hardly change" open.substack.com/pub/dominiccum…

"You’re better off having the paid versions of Opus or GPT work for you than ~99% of MPs."

@henrywinter Not walking over the crest is play acting for the cameras. Also hardly like Atletico were respectful yesterday evening.

Ben White clearly made an innocent mistake when walking across the Atletico Madrid crest. He's not showing a lack of respect. He’s caught up in the emotion of the occasion. Simeone father and especially son were unimpressed… #AFC 1/2

These are just a few examples of how we can use AI to enhance our critical thinking. Rather than just prompting "tell me about X", ask AI to challenge pre-held beliefs and assumptions, strengthen or weaken the claim. Ask how it can apply lessons from the past to your current life.

Claude on historical examples of statesman who fell foul of this rule: Prompt: Give me the clearest historical examples of statesmen who fell foul of this rule. - Bismarck himself - The bitter irony. His alliance system required him to run it. Dismissed in 1890, the machinery drifted toward WWI within a generation. - Napoleon in Russia - Predicated on a quick, decisive victory. No plan B for attrition or a Russian winter. His entire genius was built on speed and shock — when Russia simply refused to be shocked, everything collapsed. - Chamberlain at Munich - Genuinely believed he had solved Hitler with one agreement. Left zero room for the possibility that Hitler's ambitions were not finite. - The Schlieffen Plan - Knock out France in 40 days, pivot to Russia. So precisely scheduled it had zero tolerance for the unexpected. Belgium's resistance and British entry were only "unforeseeable" because the planners had refused to foresee them.

A statesman who leaves no room for the unforeseeable in history may, however, mortgage the future of his country. Henry Kissinger. I prompted Claude to evaluate this statement: 🧵

- A long period of stability creates the illusion that change must necessarily take the form of a modification of the existing framework, and cannot involve its overthrow. - A society that must produce a great man in each generation to maintain its domestic or international position will doom itself; for the appearance, and even more, the recognition, of a great man are, to a large extent, fortuitous. - Institutions are designed for an average standard of performance, a high average in fortunate societies, but still a standard reducible to approximate norms. They are rarely able to accommodate genius or demoniac power. Henry Kissinger.

"People are born as revolutionaries...The accident of life decides whether one becomes a Red or a White revolutionary." Bamberger on Bismarck. The latter once said that the former was one of the very few authors who actually understood him.