Kenneth F Levy, Principal, Wealth Advisor, CFP

@chamath Muni markets seems to take it in stride

California is going bankrupt before our eyes.

@sean_a_mcclure @alex_prompter @zerohedge It makes me think of how throughout history, all people come to believe that God exists.

What the paper calls an “Artificial Hivemind” could just as easily be interpreted as the discovery of structural invariances in human knowledge. When many independent thinkers approach the same question, their answers tend to cluster around certain stable ideas. Metaphors and frameworks recur because they capture deep regularities in reality and cognition. So convergence isn’t by default surprising. It’s in fact expected. Multiple minds navigating the same space naturally land on similar structures. We know that in physics many possible configurations exist, yet systems settle into low-energy states. When it comes to knowledge, many possible ideas exist, yet cognition gravitates toward stable conceptual structures. I am not saying the study isn’t pointing out important things to consider, I’m just saying the default “things must be broken” take is exaggerated and one-sided. For all we know, this study might be revealing something quite expected; that human knowledge has intrinsic invariances. Certain ideas keep reappearing because they are structurally optimal representations of reality. In this case, LLMs are not inventing those invariances, they are surfacing them. So many people (including researchers) are enamored with the “look, AI is not really understanding” take that they miss alternative, often more rigorous, scientific explanations.

🚨 BREAKING: Researchers at UW Allen School and Stanford just ran the largest study ever on AI creative diversity. 70+ AI models were given the same open-ended questions. They all gave the same answers. They asked over 70 different LLMs the exact same open-ended questions. "Write a poem about time." "Suggest startup ideas." "Give me life advice." Questions where there is no single right answer. Questions where 10 different humans would give you 10 completely different responses. Instead, 70+ models from every major AI company converged on almost identical outputs. Different architectures. Different training data. Different companies. Same ideas. Same structures. Same metaphors. They named this phenomenon the "Artificial Hivemind." And the paper won the NeurIPS 2025 Best Paper Award, which is the highest recognition in AI research, handed to a small number of papers out of thousands of submissions. This is not a blog post or a hot take. This is award-winning, peer-reviewed science confirming something massive is broken. The team built a dataset called Infinity-Chat with 26,000 real-world, open-ended queries and over 31,000 human preference annotations. Not toy benchmarks. Not math problems. Real questions people actually ask chatbots every single day, organized into 6 categories and 17 subcategories covering creative writing, brainstorming, speculative scenarios, and more. They ran all of these across 70+ open and closed-source models and measured the diversity of what came back. Two findings hit hard. First, intra-model repetition. Ask the same model the same open-ended question five times and you get almost the same answer five times. The "creativity" you think you're getting is the same output wearing a slightly different outfit. You ask ChatGPT, Claude, or Gemini to write you a poem about time and you keep getting the same river metaphor, the same hourglass imagery, the same reflection on mortality. Over and over. The model isn't thinking. It's defaulting to whatever scored highest during alignment training. Second, and this is the one that should really alarm you, inter-model homogeneity. Ask GPT, Claude, Gemini, DeepSeek, Qwen, Llama, and dozens of other models the same creative question, and they all converge on strikingly similar responses. These are models built by completely different companies with different architectures and different training pipelines. They should be producing wildly different outputs. They're not. 70+ models all thinking inside the same invisible box, producing the same safe, consensus-approved content that blends together into one indistinguishable voice. So why is this happening? The researchers point directly at RLHF and current alignment techniques. The process we use to make AI "helpful and harmless" is also making it generic and boring. When every model gets trained to optimize for human preference scores, and those preference datasets converge on a narrow definition of what "good" looks like, every model learns to produce the same safe, agreeable output. The weird answers get penalized. The original takes get shaved off. The genuinely creative responses get killed during training because they didn't match what the average annotator rated highly. And it gets even worse. The study found that reward models and LLM-as-judge systems are actively miscalibrated when evaluating diverse outputs. When a response is genuinely different from the mainstream but still high quality, these automated systems rate it LOWER. The very tools we built to evaluate AI quality are punishing originality and rewarding sameness. Think about what this means if you use AI for brainstorming, content creation, business strategy, or literally any task where you need multiple perspectives. You're getting the illusion of diversity, not the real thing. You ask for 10 startup ideas and you get 10 variations of the same 3 ideas the model learned were "safe" during training. You ask for creative writing and you get the same therapeutic, perfectly balanced, utterly forgettable tone that every other model gives. The researchers flagged direct implications for AI in science, medicine, education, and decision support, all domains where diverse reasoning is not a nice-to-have but a requirement. Correlated errors across models means if one AI gets something wrong, they might ALL get it wrong the same way. Shared blind spots at massive scale. And the long-term risk is even scarier. If billions of people interact with AI systems that all think identically, and those interactions shape how people write, brainstorm, and make decisions every day, we risk a slow, invisible homogenization of human thought itself. Not because AI replaced creativity. Because it quietly narrowed what we were exposed to until we all started thinking the same way too. Here's what you can actually do about it right now: → Stop accepting first-draft AI output as creative or diverse. If you need 10 ideas, generate 30 and throw away the obvious ones → Use temperature and sampling parameters aggressively to push models out of their comfort zone → Cross-reference multiple models AND multiple prompting strategies, because same model with different prompts often beats different models with the same prompt → Add constraints that force novelty like "give me ideas that a traditional investor would hate" instead of "give me creative ideas" → Use structured prompting techniques like Verbalized Sampling to force the model to explore low-probability outputs instead of defaulting to consensus → Layer your own taste and judgment on top of everything AI gives you. The model gets you raw material. Your weirdness and experience make it original This paper puts hard data behind something a lot of us have been feeling for a while. AI is getting more capable and more homogeneous at the same time. The models are smarter, but they're all smart in the exact same way. The Artificial Hivemind is not a bug in one model. It's a systemic feature of how the entire industry builds, aligns, and evaluates language models right now. The fix requires rethinking alignment itself, moving toward what the researchers call "pluralistic alignment" where models get rewarded for producing diverse distributions of valid answers instead of collapsing to a single consensus mode. Until that happens, your best defense is awareness and better prompting.

@bgurley Enjoyed the interview!

Today is launch day. 🚀🚀🚀 We find ourselves in a time with much fear/anxiety about careers. I hope this book can be a positive antidote. The permission, motivation, and methodology to do what you truly love. a.co/d/02q0aD2N

@JTLonsdale @chamath @JamesPPinkerton If $1 trillion of stock value leaves that could be zero lost revenue if the billionaires are just borrowing against it, assuming no dividend income.

@chamath @JamesPPinkerton Yep

With Zuck’s move to Florida, California’s total taxable wealth from billionaires has plummeted to well under $1T from over $2T just a few weeks ago. The loss of this tax revenue was totally avoidable but is now forever. All because Gavin Newsom stood motionless as this stupidly written bill, from a fringe union and a handful of socialist academics with an axe to grind, meandered its way into the public conversation without any action from him and freaked everyone out. These were all people that were paying 13%+ in state income tax every year WITH NO COMPLAINTS UNTIL A FEW WEEKS AGO. And now, for the rest of time, the lost tax revenues from these folks will have to be paid for by the middle class because they are the only group left in California large enough that you can tax to fill the hole. He’s forsaken the middle class instead of managing the budget, managing the deficit, eliminating even a portion of California’s gargantuan waste and abuse. He could have done any of these things at any point over the past 7+ years. But he was silent. And now California’s budget will implode and he wants to run for President. Insane.

@chamath Years and years in the making.

Don’t forget about the unfunded pension liabilities. California is on the hook for $1T in pension liabilities. It’s totally unclear how it pays these obligations. We need a hard reset on our approach to government in California. What we’ve been doing ISN’T WORKING!

@elonmusk It worked.

Note: With NSFW enabled, Grok is supposed allow upper body nudity of imaginary adult humans (not real ones) consistent with what can be seen in R-rated movies on Apple TV. That is the de facto standard in America. This will vary in other regions according to the laws on a country by country basis.

Can anyone actually break Grok image moderation? Reply below.

@JamesBradleyCA @theallinpod @friedberg @adamcarolla I wish. Most Californians are Democrats and want more government in our lives.

@theallinpod @friedberg @theallinpod @adamcarolla @friedberg Nailed it. Most Californians aren’t ideological warriors—they just want government out of their homes, schools, and driveways. But the state won’t let them be left alone. Time to change that.

Adam Carolla: Most people on the right “just want to be left alone.” 🔥 @friedberg: “I never was interested in politics. I hate talking about it, I hate getting involved in it.” “I’ve always just been a guy like, leave me alone, let me do my work, let me live my life.” “I don't want to think about government and politics, but it's become so prevalent in our lives.” “It's become something that all of us are almost forced to discuss.” @adamcarolla: “ Everybody I know who gets lumped into being a right winger, all they say is, ‘I just want to be left alone.’” “I have a property. It's my property. I own it. I pay taxes on it. I would like to rebuild my property.” “And the government says, ‘No, you cannot.’” “And now I have a problem. And it's getting political, but I don't want it to get political. I just want to be left alone.” “I got a niece and she's five years old, and I don't want a nine foot tall tranny reading her Cat in the Hat at the library. This is avoidable. You can just leave her alone.” “And you can leave me alone, and I'll pay my taxes, and I'll be a good neighbor, and I'll pick up my dog's poop. And we could get on with our lives.” “But you won't have it. You have to get involved.” “And most of the people I know, who are on my side of the aisle politically and just sort of emotionally, are going ‘I just want to be left alone.’” “If I want a gas stove, I want a gas stove. If I want a gas powered truck, I want a gas powered truck. And I would like to be able to rebuild my property using my money and then pay taxes on that without you getting so involved.” “So I'd just like to be left alone.” **************************** Thank you to our partner for making this happen! When you need a partner trusted by millions, there’s one platform for all business - PayPal Open. Grow today at paypal.com/us/business.

@JulianKlymochko Many factors to consider. Here is a shopping list...or just buy the whole thing all at once.

The bear market in private credit (listed BDCs) has caused the average yield to rise to 12.6% and the spread to surge to 893bps above the 5-year Treasury, up 247bps since 2021 and 216bps above the four-year average. Given widespread double-digit NAV discounts, the average listed BDC yields 290bps more than the average non-traded BDC.

Shill me a project actually building something useful please

@PalmerLuckey @unseen1_unseen Your Lord Jesus Christ? For real?

Late last night, I received a final Christmas gift: official notification from the PRC Ministry of Foreign Affairs that I, Palmer Luckey (Male), have been personally sanctioned as a dangerous actor. I want to thank my family, my team, and my Lord Jesus Christ for this award. Anduril has been sanctioned for a while now, as have many of my peers, but it means so much to finally have my non-existent Chinese assets seized and repurposed. The sanctions also prohibit all Chinese nationals from engaging with me in any way, which should really clear up my social media feeds. Merry Christmas!

@mikealfred You discussed this over a beer?

BREAKING: Bitcoin founder Satoshi Nakamoto says in a new Fox News interview that Bitcoin has been in a shadow bear market for the last 4 years and will only return to a full bull market when quantitative tightening in the United States ends later this week.

@haralabob @theallinpod Of course people should move if they can’t afford to survive in their home town.

J-Cal: "Ben Shapiro is absolutely correct." “The people who are upset at Ben Shapiro are a bunch of dipsh*t hipsters who went hundreds of thousands of dollars in debt, are paying $5000-6000 a month in their rent, can never get out from under their rent payment or their liberal arts bullsh*t degree.” “If you're a smart person, go to the University of Texas, graduate with little to no debt, live in a modest apartment, put down a down payment, and buy a $500,000 home. This problem doesn't exist in Texas. It doesn't exist in a lot of markets.”

@chamath Can personally attest to microdosing for reduced inflammation

This may convince me to start microdosing Tirzepatide. What does this drug NOT do?!

@MilaLovesJoe @unseen1_unseen The analysis by Grok AI on X said Elon was wrong?! Who do I believe? Elon of his AI?

Elon Musk details D.O.G.E.’s most SHOCKING discoveries.

@AndreasSteno What he said.

Hvorfor sætter ingen politikere fra midten/højre på Københavns Rådhus en kampagne sammen på det her? Line Barfod fra Ø bliver ved og ved og ved med at påstå, at hun kæmper for lavere boligpriser i København – men hendes politik peger i stik modsat retning, så det er jo ganske enkelt løgn. Intet mindre. Sidste år stemte Enhedslisten imod 48.000 nye boliger i 2036 – de ønskede kun 40.000. Nu er deres målsætning yderligere faldet, som grafen viser. Det er altså ikke bare en fejl – det er en strategi. Line Barfod og Enhedslisten fortsætter deres ræs mod bunden med endnu mere forgyldning af private og multinationale boligejere, mens boligpriserne presses opad. Det her burde udstille det himmelråbende hykleri hos den københavnske venstrefløj. Det er helt legitimt at ønske mindre byggeri – men så stå dog ved det og de konsekvenser det har. Graf fra @RasmusDamgaard5

@jaredkushner @realDonaldTrump And thank you!

“Thank you President @realDonaldTrump for your commitment to bring this horrible conflict to an end.”

@JTLonsdale Top quality CEOs and athletes can be compared to Bitcoin. Scarcity.

This is misguided populism and it’s wrong, and unAmerican. The CEO deserves whatever the market will pay for top talent, in a free society, as does a movie star or rock star or top basketball player, etc. Let’s get skills and wages up, but not by attacking talent and success.

@JulianKlymochko Can those NAVs be trusted? In a bad economy, those could come down in a hurry.

With the recent negative sentiment and drawdown in the liquid private credit sector, yields have risen to an average of 13.9%, matching April's peak. Currently, 80% of listed BDCs trade at a discount to NAV, with the median private credit fund trading at 81 cents on the dollar.

@Jason @MamaNouraKSA Live now. Set the rest aside: youtube.com/live/DKGRfogo-…

I went to Cava once, its was solid but not as good as @MamaNouraKSA

@ericjackson Live now: youtube.com/live/DKGRfogo-…

Ironically I will be going to Germany this week, just as Drake is going back to Canada!