Chamath: 3 Things Every Company Needs in the Age of Super Intelligence 1) End-to-End Traceability 2) Policy-to-Risk Mapping 3) Auditable Evidence @Chamath: “There's really three things that you're going to need in an age of super intelligence so that boards are protected, CEOs have visibility, and all of the other compliance infrastructure and governance infrastructure of a company like D&O Insurance and all these other things are kept in place, which is end-to-end traceability, right? Which is to say, when you deploy super intelligence, you need to know what's going on. You need to map it to the policies so that you understand what the risks are. All companies have standard operating procedures, but you need to know exactly when you're deploying super intelligence, how that builds on top of your standard operating procedure, so you need to map policies to risk. And then the third is you need to have auditable evidence, which is your way of showing all of your downstream ecosystem, ‘Here's what I've done.’ So if anybody taps you on the shoulder, whether it's a lawyer, whether it's an auditor, whether it's a regulator, whether it's in a lawsuit, whether it's a customer, ‘Hey, how did this happen?’ You can present something. So these are, like, very practical, sane things, and you have this solution where the alternative was to hit the pause on trillions of dollars of GDP so that we could sort this out in a multinational body. The alternative was crazy, so I think what happened was historic and important.”

POD UP! 🚨 -- Inside Trump’s Super Intelligence Summit -- How the AI Safety Accord Happened -- Data Center NatSec + Cybersecurity Arms Race -- GDP Beats + Other Economic Data -- Midterm Predictions -- Media Coverage of Attempted Plane Hijacking (0:00) Bestie intros! (2:01) Trump’s Super Intelligence Summit: Behind the scenes and how the accord was created and signed (17:57) Cybersecurity arms race, data center national security, America.gov, coordinated doomer messaging (45:22) Recent economic data: GDP, jobs, diesel, inflation, rates, manufacturing, poverty, and more (1:00:24) Midterm predictions, 2028 look ahead (1:09:28) Media coverage of attempted terrorist hijacking

We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks: Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better: Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better: Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better: Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work! In summary: - As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding. - Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.

.@altcap on @POTUS's SI Summit: "@POTUS set the tone at the start of the meeting — making sure that this is about main street Americans." "It was a real legendary move of statecraft."

This is DoorDash Air. As a pilot, I can tell you that you wouldn’t build a plane and then ask if it can carry cargo. So we built @DoorDash Air from the ground up, designed around what 13 years of deliveries have taught us: what people order, how much it weighs, and how far it has to travel. Our drones are as silent as a passing car and early tests show that we’ve gotten flight delivery times in under ~5 mins on average, so you can get your @chipotletweets & @popeyes (our first partners) and local favorites right to you, fast. The first deliveries will be in NorCal. We proved autonomy works on the ground. Now see how it works in the sky.

Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry. Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come. But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility. This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems. Together, we are building the foundation of the AI economy. Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. nvda.ws/4hcoq7m

A timeless rule of competition: commoditize your complement. Every product has complements. When a complement gets cheaper, demand for your product can rise. So companies often try to drive the price of key complements toward zero. Open-weighting a model is basically this move at a frontier scale. OpenAI and Anthropic position the AI model itself as the product you pay for. For Meta, dependence on a model layer controlled by competitors would threaten its ability to profit from its own AI products. So Meta gave Llama for free. They can afford to give away the model because they don’t have to sell it. Instead, they sell advertising to 3B users. The model is Meta's complement, while its data and distribution are the moat. Once you start to see the pattern, other open-model releases become more clear: - Google (Gemma) and Microsoft (Phi) sell cloud computing and software. A cheap model layer sends more workloads to their clouds. - OpenAI (gpt-oss): When open competitors set the price, you release something open too. - Nvidia agreed to acquire Hugging Face for $12.9B, preserving openness and support for competing hardware and cloud providers. Making models more accessible encourages more applications and greater usage, which expands demand for computing infrastructure. Essentially, companies open up one layer when doing so strengthens the part of the stack where they make money. For companies whose only business is selling access to closed frontier models, cheaper and more capable open-weight models mean more competition at the model layer and more pressure on token pricing.

#3 on Apple Podcasts

Chamath on regulatory risk impacting Anthropic's IPO: "It's probably going to clear at a much lower price than anybody thinks." "You have to throw everything in the kitchen sink into disclosures. The problem is that that will make what is an otherwise dense and impenetrable S-1 even more dense and impenetrable." "You can superficially cover your regulatory risk. But the way that it actually gets translated in the market is you're just going to have to water down the expectations of the IPO sellers and create a much larger margin of safety for the IPO buyer." "Even if you think that they're on a $100 billion run rate, what would have been a $2 trillion market cap, now you're at $1 trillion or less." "The hedge funds and the pension systems and the long only mutual fund complexes who have a fiduciary responsibility to their sources of capital will justify all the risks by essentially asking for a lower and lower price." "I think you're going to see a much more moderate and reasonable IPO motion here. It's probably going to clear at a much lower price than anybody thinks." "I think it's good for Anthropic because it clears all this noise. It forces them to just be a corporation and then run the business. And I think that they'll do great. But these other risks are going to weigh on the IPO. And the only way to flush them is to just bring the price way down."

POD UP: The Besties are Back! 🚨🔥 -- Anthropic IPO at Risk, OpenAI's Pushed Back to 2027 Over "Safety" -- AI Chaos Reactions, The Frontier "Lab" Problem -- Token Prices Fall, Open Source Gains Share -- Political Reactions to Pacing the Frontier: Trump, Obama, Bernie, Bessent -- Meta’s Muse Moment -- Has AI Alignment Backfired? (0:00) Bestie Intros: Favorite All-In Summit moments (3:54) Reacting to AI chaos: Liability, competition, and rebranding frontier "labs" (20:03) Open source performance and cost: What this means for frontier companies (29:22) Anthropic and OpenAI postpone IPOs: liquidity risk and open source pressure (53:11) Political reactions to "Pacing the Frontier": Bernie's AI ban, Trump, Bessent, Obama (1:07:58) Meta launches Muse, AI "alignment," Oracle's Force Majeure (1:27:13) Anthropic’s bio research and "wet lab" in SF

Chamath on Obama’s AI Warning: Follow the Money President Obama last week: “If we are thinking about AI just in terms of how do we cure cancer or get better energy, you can do that without having agentic AI. That's a misalignment between what our society needs and the commercial imperatives that these companies are facing.” @Chamath: “This is a very important moment for a very simple reason, which is we, as in the world, are about to endow three, four, five, six companies with about $10T of wealth. And what Obama knows very well is that most of those companies are overwhelmingly left-leaning. And what he also knows is that there is a huge portion of that money that will then get put into philanthropic and charitable causes that then he, and the people around him, will be beneficiaries of. That is the truth. We already know this because we know that some of these frontier corporations actually ask you to stand up DAFs and have a portion of your stock that you're willing to pledge. So this money is going to go to things other than consumption or savings. And so it's going to go into PACs, it's going to go into political movements, and (Democrats) stand to disproportionately benefit. So this has nothing to do with prosperity. This is a very simple political calculus. If you freeze frame the economy the way it is today, a handful of organizations that will disproportionately be able to affect the Democrats will win. They will capture the lion's share of the economic gains, and then they will help the Democrats win power. That's all this is.”

“hey muse, can you get the gulfstream in position for me to go to my second fitting in milan?” muse: