The Ledger
Every statement that passed quotation and attribution checks. Mix any filter with any other: certainty 1/5, debate potential 5/5, or both at once.
why aren't all 46 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 0 checkable ones are still open, waiting for their date. predictions held up or didn't; assertions are supported or contradicted. on every card: ▮▮▮▮▮ certainty · ▮▮▮▮▮ debate potential. speakers are clickable
Schneider: Use LLMs to write CPU software, don't burn tokens running tasks
“You should not be paying Anthropic. You should not be paying Chad CPT to do an API call. You should be paying them to make the software that uses CPU to do the API call. Why are you paying this tax on tokens every time that you're trying to do this marketing a…”
Shipper predicts OpenAI and Anthropic launch agent app stores within a year
“And I bet you within a year the opening eyes in the anthropics of the world will have app stores for apps that are built specifically to be used inside of these services. Maybe even cursor too. I don't have any insight on any of this.”
US government letter forced Anthropic to shut down Claude Fable 5
“Then Friday at five 21 PM, the US government sent Anthropic a letter. And by Friday night, the model was gone. Disabled.”
Wilkinson: Anthropic and OpenAI will soon launch AI CEO products
“My, I mean, my kind of belief is like pretty soon Anthropic and OpenAI are gonna launch like basically like CEOs. So you just be like, I'm just gonna hire a CEO for my company. Give me your data. I'll figure it out. I'll run it. I'll run the whole business.”
Liu: OpenAI and Anthropic have over $80 billion in combined revenue
“Take opening Ionanthropic alone, right? Let's just say they have a combined revenue probably of like eighty million plus, right? Or eighty billion, sorry, plus right now up from like basically zero a few years ago.”
Brown: Codex beats Claude Desktop by unifying coding and coworking tools
“Another way that I like to think about this is Claude code or this codex is like Claude code and co-work in a single product, but it's even better than that. And the reason why I like it more than Claude desktop is Claude made a decision to split up co-work an…”
Moritz Kremb: OpenAI permits OpenClaw OAuth connections while Anthropic officially bans them
“Because OpenAI has stated that they're okay with it. So for them, it's definitely fine to use this OAuth method with Anthropic. It's kind of a gray area a bit. I think officially in their terms of service, it says you're not allowed to do that.”
Moritz Kremb: Anthropic is banning users for connecting OpenClaw via OAuth
“And some people have been getting banned over it. So I do recommend that if you, if you're scared about your account getting banned, just create a new one, create a new account and put the 20 dollar plan on that.”
Ross Mike: LLM coding agent performance degrades above 50% context limit
“I generally wouldn't go over 50%, meaning like the anthropic model Opus 4.5 has a 200,000 token context limit. The moment, in my opinion, you've got over a 100,000 tokens, meaning you're using the same session. It starts to sort of deteriorate. That's when you…”
Ross Mike says he would cancel Claude subscription for Astra
“Anthropic is sweating for sure because for the first time I would like, if I, if someone asked me pick a subscription, I would cancel and drop my Anthropics subscription. I would cancel Claude. Astra's that good.”
Shipper: Anthropic's Claude and Opus remain ahead in visual product design
“I would go to Claude for the actual design, and then port it back in here, because I still think that Claude has a better sense, especially Fable, RIP. But even, even four eight would be better at this. And I think for a product like this, being able to see it…”
Shipper: OpenAI and Anthropic do not really offer fine-tuning anymore
“In that case, like, you have to use open-source models, because none of the major model companies, or at least OpenAnthropic, don't really do fine-tuning anymore.”
Dan Shipper: Anthropic holds 'mandate of heaven' while OpenAI is underrated
“I feel like Anthropic has the, has a little bit of the mandate of heaven right now, and OpenAI is quite underrated.”
Isenberg: OpenAI or Anthropic should acquire note-taking app Obsidian
“If I'm open AI or anthropic, I'm buying obsidian.”
Autonomous AI agents will massively multiply token consumption and revenue for Anthropic
“I think that's one of the very good things for like, Investors in Anthropic, right, is with agents and agents now being, I think probably the new killer feature in Opus. You're going to take whatever token usage and multiply it by the number of agents.”
Isenberg: Claude 3.7 Sonnet is the best writing LLM available
“The best writing right now, as of recording, this is the best writing LLM that exists.”
Kahl: OpenAI and Anthropic are structurally constrained from optimizing niche tasks
“Like OpenAI, Anthropic, all of these players, they have to have models available that can do everything well. So they can't really have models that do one thing specifically, specifically well, right?”
Coogan: Stolen AI Model Weights Are Far More Usable Than Web Codebases
“If you stole the code base to Facebook, And you were like, okay, I have all the code to Facebook. I'm going to launch Facebook to.com. Well, you don't have any of the users. People aren't going to your website. It's not valuable, but with the large language mo…”
Isenberg: Claude is currently 30% better than ChatGPT
“No it's honestly, 30% better. Like, every answer you get is 30% better. So, the thing is, ChatGPT was better, and then Clode recently caught up. Now ChatGPT is getting better, but right now, as we're recording this, Clode is where you want to be.”
Why Anthropic's Claude Is a Weak Brand Name in America
“Claude.ai is, you know, on one hand is good because it's like a person and a brand, and you can kind of see it, and it's French, so maybe it's like, I don't know, higher quality. But it's hard to spell for the, for Americans. They don't really get it, you know…”
Isenberg: Claude Code Requires the Same Scaffolding as Human Employees
“If you want to, if you want Claude Code to act more like an employee, you need to give it the same basic things you would give a person joining your company.”
Isenberg: Agent founders should start with workflows and earn autonomy
“Many agent problems should start actually as workflows. So a workflow follows a predictable path. An agent decides more dynamically. Founders should earn autonomy by starting with the predictable path and adding judgment only when it creates value.”
Brown: Claude and Codex Subscriptions Provide Thousands in Subsidized Tokens
“Like, in Claude or in Codex is you just save so much money because they're subsidizing, like, you get thousands of dollars of tokens for your a hundred dollar per month or 200 dollar per month plan because you're using their platforms.”
Isenberg: Claude Design output delivers thousands of dollars in value
“Like the value right now has been thousands of dollars. Period. The value has been thousands of dollars. Did you see that deck? That deck was crazy. The directions, the wireframe directions were on point, and the visual style was really good. It nailed 90% of …”
Isenberg: Claude Design video capabilities are a 5/10 at best
“I think the video capabilities of Claude Design is a five on 10 at best.”
Isenberg: Claude Design offers the best wireframing seen to date
“I think it's, like I said, it's the best wireframing I've seen to date.”
Moritz Kremb: Claude Cowork is just Claude Code with a better UI
“So they kind of built cloud co-work and put that in the app with just a nicer UI. But under the hood, it's basically the same as Cloud Code.”
Isenberg: Founders over-optimize debating OpenAI versus Anthropic frontier models
“You don't need to worry about, you know, is OpenAI a little bit better than Anthropic? Like, the reality is, you know, you can think of it as like Ferrari and Lamborghini, like both cars are going to go fast. And, you know, one day one car might go faster, and…”
Claude Opus 4.6 creates agent teams while Codex requires direct reasoning prompts
“When I'm talking to Opus, I can tell Opus, build me a team, and here's what I want each member of the team to do. When I'm talking to Codex, I can't really tell it to build me a team, but I can tell it to think about stuff.”
Ross Mike: Autonomous coding loops without clear PRDs waste token spend
“If you have a terrible plan, if you have a terrible PRD, This doesn't matter. You're just donating money to Anthropic, and I wish you the best of luck if that's what you want to do. But if you want to make sure that your tokens are not wasted, you're going to …”
Ross Mike: Manual iteration with Claude Code builds better product intuition
“And this is why a lot of people who were fighting with Claude code all these months are really, really good at using it now because they spent the time building without using these crazy automation loops.”
Isenberg: Leaked Claude Agent Mode Is a Huge Leap for AI
“So I think that this is a huge leap
for AI in general.”
Isenberg: AI Domain Terminology Triggers Sophisticated Reasoning in LLMs
“Using advanced prompting terms can trigger more sophisticated modes of operation. So models are trained on a vast amount of text about AI itself. So using terms from the field activates specific powerful behaviors.”
The Boring Marketer: Claude Opus 4.5 One-Shots Interfaces With Front-End Skill
“4.5 Opus, I think is going to be an incredible model just based off some of these initial tests that we did and that I was doing last night. When you combine it with that front end design skill, it's able to one shot some pretty incredible interfaces and desig…”
James: Claude Opus 4.5 will resolve bugs with less sub-agent prompting
“I'm thinking that its ability to,
Work through the bugs without sort of getting lost is going to be a lot stronger.
and I wouldn't maybe need ultra think or sub agents or maybe even those additional agents on the side.”
Riley Brown: Claude Sonnet and Opus are agents out of the box
“Claude Sonnet IV and Claude Opus IV are literally agents out of the box, and so what we're talking about is giving tools to agents.”
Anthropic played 3D chess by pushing integration maintenance onto service providers
“This is why I think Anthropic, they were playing three D chess when they built this, is the way this is architected, the MCP server Is now in the hands of the service provider.”
Brown: Vercel's v0 is like Claude Artifacts on steroids
“And so this is a lot like clawed artifacts, but like on steroids.”
Brown: GPT-5.5 API Costs 20% More Than Claude Opus 4.7
“And then it's also 20% more expensive than Opus 4.7.”
Anthropic used Remotion to generate early product launch videos
“Anthropic did this when they first started doing their product launches. They were just launching them With ReMotion videos.”
Gaskill: Anthropic built MCP as a universal tool translator
“Anthropic built MCP to basically sit as this translator in between your tools, so that Claude can still just speak English, and your tools can just speak their languages, and this MCP speaks every language, and then just translates your calls from your agent t…”
Altman announced GPT-5.3 Codex 18 minutes after Anthropic released Opus 4.6
“Opus four, six came out and then Sam Altman put together a quick tweet. I want to say like maybe 18 minutes later announcing GPT five, three codex”
Isenberg: OpenAI Adopted Anthropic's Agent Skills Standard for Codex
“So what's really cool is they're just following the agent skills IO standard, the one by Anthropic.”
Cheung: The Rundown drafts newsletters manually, using Claude as editor-in-chief
“When we write our newsletter, we write it all by hand, But we like to use Claude as like an editor in chief.”
Anthropic's Markdown choice makes Claude Skills accessible to non-technical users
“Markdown files are super easy to understand. If you're non-technical, Obviously you can really understand it, so I'm happy that they, you know, picked such an easy way to actually, ah, to get involved.”
Isenberg: Break Complex AI Prompts Into Step-by-Step Subtasks and Synthesis
“For a complex task, act as the conductor. Prompt for each part separately, then prompt for the synthesis. So don't ask for a ten-page report in a single prompt. Break it down into logical subtasks that you manage step by step.”