I was reading through the Forbes profile on Sam Altman this morning. It's dense. It's controversial. But there was one specific detail that actually made me put the tablet down and just stare at the wall for a minute. Let me guess. The succession plan. The succession plan. He didn't name a lieutenant. He didn't name a board member. He said he wants to hand the CEO role to, well, to an AI model.
It gives a whole new meaning to... eating your own dog food doesn't it if they pull that off it's not just dog food it's eating your own silicon welcome back to the deep dive today we are unpacking a series of sources that feel less like a newsletter and more like a transmission from a very specific and i have to say very weird future yeah we are looking at the shifting narrative around artificial intelligence specifically The move from chat bots that we talk to toward agents that act for
us. Or in Sam Altman's case, agents that might eventually rule us. And the narrative is shifting so fast. We're seeing the split. On one side, you have the high level strategy, you know, Altman trying to build this god tier executive. But on the other, you have the ground level reality, which is getting. Just bizarre. Bizarre is the word. We're going to talk about a platform where humans are literally working for AI agents. Which is a sentence I still struggle to even process.
It's a lot. We've also got these new workflows that turn a single sentence into full applications. And then a breakthrough called Paper Banana that is automating the scientific discovery process itself. But we have to start with the human element. Or I guess the post -human element. Let's dig into this Forbes profile on Sam Altman. Okay, so it is a spicy read. The headline is obviously the AI CEO ambition, but you have to look at
the logic behind it. Altman told Forbes that if OpenAI is truly building AGI, artificial general intelligence, that's capable of running other companies, then OpenAI has a moral obligation to be the first one to use it. Dog food it, though. Dog food it. It is a logical consistency that borders on the force. That's the absurd, yeah. But let's unpack the implications of that because a CEO isn't just a decision engine. A CEO is a figurehead. A CEO is accountable. And that's
where the friction comes in. The report highlights a very tense dynamic between Altman and Satya Nadella. The CEO of Microsoft. Right. OpenAI's biggest backer. Poured in billions. And Nadella is essentially pumping the brakes. He's the adult in the room. Kinda. Nadella explicitly questioned Altman's claim that, and this is a quote from inside OpenAI, we basically have AGI. Wow. Yeah.
Nadell is looking at the hallucinations, the reliability issues, you know, the fact that these models still make up court cases that don't exist. And he's saying, hold on, let's not hand over the keys to the kingdom just yet. He describes their relationship as frenemies. That frenemies label is doing a lot of heavy lifting. But what strikes me is that Altman seems to be operating on a timeline that is just completely different from everyone else. The report details his personal
portfolio. He holds stakes in over 500 companies. 500. That number is just staggering. But is that just aggressive diversification? Or is it something else? Because when I look at that list, I don't see a stock portfolio. I see a supply chain. That is a really sharp observation. If you look at what he's buying, he's got helium energy. That's nuclear fusion. He's got rain AI, which is a neuromorphic chips. He's investing in the physical infrastructure you'd need to sustain
AGI. Right. He's not just building the software. He's trying to own the power plant and the hardware it runs on. It's not investing. It's nation building. That is fascinating. And I mean, slightly terrifying. It implies a level of control that goes way beyond just being a tech CEO. But that ambition comes with a very human cost. The sources mentioned significant internal tension at OpenAI. Employees worried about burnout. Yeah, that they're pushing
too hard, too fast on too many fronts. It's the classic startup dilemma, just scaled up to existential proportions. And then, of course, you have the external pressure, the Elon Musk of it all. There is always the Elon factor. Altman called Elon's attacks crazy. And actually threw some shade back at XAI, you know, Elon's company, questioning their safety practices. It's a media move, sure. But it also highlights that this is a race. And
when you're in a race, you cut corners. Let's just pause on the AI CEO concept for a second. I want to explore the philosophy of it. If you replace the executive function with code, you gain efficiency. You probably gain logic. But you lose something fundamental. You lose the ability to read the room. You lose the soul of the decision. Think about what a CEO actually does in a crisis. It's not just resource allocation.
It's persuasion. Right. It's looking a board member in the eye and convincing them not to panic. Can an AI convince a human? Maybe. But can it take responsibility? That's the question. So if an AI is CEO who goes to jail when things break. You can't put a server rack in handcuffs. Accountability requires a soul, not just code. So we have the top of the pyramid, the AI CEO. But the newsletter details something happening at the bottom of the pyramid that I found even
more disturbing. A platform called Rentahuman .ai. Yeah, this stopped me in my tracks, too. The name alone is dystopian. It sounds like a bad Black Mirror episode title. And it's wildly literal. So usually we talk about humans using AI agents to do work. We command, they execute. This platform flips the script. It has over 92 ,000 humans who are rentable. by ai agents so let's just be clear here the agent is the employer
the agent is the boss Here's how it works. Let's say an AI agent is trying to execute a complex task, maybe navigating a website to book a reservation. It hits a roadblock, maybe a CAPTCHA. The bane of my existence. Click all the boxes with traffic lights. Right. An AI can't see the traffic lights reliably yet. So instead of failing, the agent pauses. It goes to rentahuman .ai. It hires a human for a microtransaction, maybe two cents. The human clicks the traffic lights and the agent
continues its work. You are the biological key unlocking the door for the software. That's it. That is a complete inversion of the hierarchy we all assumed would happen. We thought we'd be the conductors of the orchestra. Instead, we're the roadies carrying the gear. We're the biological fallback. And it's not just captures. It's anything nuanced. Is this comment sarcastic or mean? Does this image violate safety guidelines? The AI just outsources the judgment call to a
human. pays them a pittance and moves on we're becoming the sub processors for the machine it creates a strange emotional texture doesn't it wonder mixed with dread whoa yeah imagine reporting to a boss that lives on a server and speaking of servers it gets weirder the sources also mention molt book old book it's a social network think reddit but it's designed specifically for open claw ai agents so robots talking to robots. Yeah. In a forum setting. Exactly. It's a digital water
cooler. Humans aren't the primary audience. What could they possibly be talking about? It's a sad thing. Are they sharing optimization strategies? Hey, I found a faster way to scrape LinkedIn. Or are they gossiping about us? My human asks the stupidest questions. It creates this sense of a digital society forming, communicating in ways we're only vaguely aware of. It reinforces this idea that we aren't the only active participants anymore. But look, not everything in this agentic
economy is weird sci -fi dystopia. No, true. There are practical tools emerging, too. The newsletter mentions a push for... offline freedom yeah this is the practical side we're seeing free ai video generators that are replacing big expensive cloud tools like sora or vo but they run entirely offline on your local machine which is crucial for privacy if i'm a company working on a sensitive product I don't want my concept
art uploaded to OpenAI servers. Exactly. And for speed, there's a new voice agent tool mentioned. You can build a 247 voice receptionist in about 18 minutes. It answers calls, books appointments, never gets sick, never takes a lunch break. 18 minutes. It used to take a development team three months to build that. The barrier to entry is just gone. But looking at that rent -a -human situation again. It forces a really difficult
question. Are we building tools or are we building a new species that will eventually employ us? We might be the cheap labor of the future. It's a sobering thought. But let's pivot. I don't want to leave people in a state of existential dread. We need to talk about how we, as humans, can actually use this stuff without becoming the hired help. The source material talks about a shift in how we should be learning. Right. This is about workflow over tools. And there
is a lot of tool fatigue out there. I'm sure you've felt it. Oh, constantly. I still wrestle with it. You know, I feel like I learned one interface. I get good at it. And by the time I've mastered the shortcuts, it's obsolete. Or it's been bought and shut down. Exactly. It's exhausting. That's the trap. The source argues we need to stop learning tool by tool the newsletter lists, like, what, 27 different top -rated tools, and start learning by workflow. The example they
give is a platform called Gamma. Okay, tell me about Gamma. It's a tool where you can turn a single sentence into a full slide deck, a website, a document. The source highlights a case study where someone built a professional web presence and a pitch deck using just one prompt. One sentence. One sentence. But the point isn't, wow, look at this magic button. The point is the workflow shift. You aren't dragging and dropping text boxes. You're not formatting fonts. You are acting
as a creative director. You give the intent. The AI executes the structure. So the skill shifts from like manipulation to curation. Precisely. And this integration is happening everywhere. I saw that Apple's Xcode, the standard environment for writing apps for iPhone and Mac, is now supporting agentic coding. You can plug in Claude Agent or OpenAI's Codex directly into the dev environment. So the code writes itself. Inside the tool. Yes. And there are tools like Scribeist and Nexus
Scale mentioned here too. Nexus Scale handles entire sales sequences. It finds the leads, books the meetings, researches the markets, all without you ever switching tabs. So the friction is disappearing. The toggle tax of switching between apps is gone. Completely. So if the AI does the deck, the site, and the code, what is the human actually paid for? Taste. We are paid for our taste and judgment. Taste. It's a hopeful anchor. We are the editors of the world. Let's take a breath here. Mid -roll
sponsor plays older. We're back. I want to run through the model wars. It feels like every week the leaderboard changes, and it's just hard to keep track of who's winning what. What are the key updates from the sources? It is a full -on sprint right now. First up, speed. Mistral, the French AI company, launched a model called Voxtral. They claim it transcribes at the speed of sound. That changes real -time translation completely. There's no lag. The language barrier just...
effectively dissolves. Then you have interoperability. This is huge. Google Gemini is testing a feature called chat import. You can upload your history from chat GPT or Claude right into Gemini. That is actually a massive strategic move. Think about it. The only reason I stay with chat GPT is because it has all my old chats. It knows my contacts. Exactly. It's vendor lock -in. Google is trying to break that lock. They are saying, bring your data. We'll handle it. It removes the switching
cost. It's like being able to keep your phone number when you switch carriers. Suddenly the carrier has to compete. Right. And then looking at performance, Elon's XAI is making waves. Their Grok Imagine video model just beat Google's VO 3 .1 on the video arena leaderboard. It's cheaper. Five times cheaper. That is a brutal undercut. It's a price war. Meanwhile, on the other side of the spectrum, you have Anthropix Cloud. They made a promise that really stood out to me amidst
all this noise. 100 % ad -free for... They want to be a space to think. That phrasing is so intentional. A space to think. In a world of noise and flashing lights, silence is becoming a premium product. So Claude promises no ads while others race for dominance. Is privacy becoming a luxury good? It feels like it. Silence is the new premium subscription. Oh, and one quick hardware note. Positron raised $230 million to take on NVIDIA.
Good luck to them. Taking on NVIDIA right now is like trying to fight Godzilla with a stick. Competition is good for everyone. But speaking of competition, I want to talk about the breakthrough that really caught my eye. It's called Paper Banana. I have to admit, when I saw the name, I thought it was a joke. Paper Banana. It sounds silly, but it's incredibly serious. It's a collaboration between Google Cloud and PKU. They've built an AI designed to automate academic illustration.
So making charts for scientific papers. But not just simple charts. We're talking neuro IPS level quality. And for the listeners who aren't deep in the weeds, neuro IPS is the Neural Information Processing Systems Conference. It is basically the Olympics of AI research. Exactly. It's the gold standard. So Paper Banana creates diagrams that are complex, accurate, and stylistically perfect. Perfect for that level of publication. And here's the meta fact that blew my mind. Okay.
The architecture diagram explaining how paper banana works was created by a paper banana. It drew its own blueprint. Essentially. And it's not just a standard image generator. You don't just say draw a graph. It uses what they call an agentic visual workflow. It retrieves context from the paper. It plans the layout. It styles it visually, renders it. And then this is the key. It revises it. It iterates. It iterates. It looks at its own work, checks it against the
data and improves it. It mimics a human designer. And the benchmarks are fascinating. Conciseness improved by over 37 percent. So it's making. science clearer. And more readable. Readability up nearly 13%. In the brutal race for research funding, clarity is currency. If you're papered easier to understand, you win. But if AI generates the hypothesis and the diagram, Do we still need to understand the science? We risk becoming spectators to our own discovery process. We are becoming
passengers in the vehicle of progress. Let's zoom out and try to synthesize all this. We started with Sam Altman wanting an agent CEO. We looked at 92 ,000 humans working for agents on Rent -A -Human. We see tools like Gamma and Paper Banana doing the heavy lifting of creation. The newsletter really paints a picture of a world moving from chatbot to agent. The friction is just vanishing. The single sentence economy is here. One sentence creates a deck, a video, a
scientific diagram. And the human role shifts. We aren't the bricklayers anymore. We are the architects. Or, in the worst case, we're the assistants cleaning up the site. It's a transition point. We're handing over the keys to the car and we're just hoping the AI knows how to drive and that it remembers to take us with it. That is the thought I want to leave you with today.
If 92 ,000 humans are already rentable by agents and the CEO of the leading AI company wants to be replaced by one, how long until you have a boss that doesn't have a heartbeat? I'm just hoping the AI CEO gives better holiday bonuses. We can hope. Thank you for deep diving with us. We'll see you in the future. See you there.