mirrored (@mirrorxbt) header
mirrored (@mirrorxbt) profile picture

ai/sI maxxing - no life

not found
Joined August 2026
Photo by mirrored (@mirrorxbt)Photo by mirrored (@mirrorxbt)Photo by mirrored (@mirrorxbt)Photo by mirrored (@mirrorxbt)Photo by mirrored (@mirrorxbt)Photo by mirrored (@mirrorxbt)Photo by mirrored (@mirrorxbt)Photo by mirrored (@mirrorxbt)Photo by mirrored (@mirrorxbt)Photo by mirrored (@mirrorxbt)
This is f*cking insane. Chamath Palihapitiya, the billionaire who ran growth at early Facebook, just said software stopped being worth anything this week: "I think that IP and software was effectively rendered worthless two days ago When everything is headless, and you can wrap them in agent tools, and you can connect them via MCP to other experiences, the value of IP does no longer exist in software" He says people decompiled Adobe's 5 core products and published open-source versions. Moats that took decades, gone in days In 2 minutes, he explains why the software business just flipped Paid Software → Headless Services → Agent Tools → Software Worth $0 Headless means the app doesn't need a screen anymore, an agent calls it directly MCP is the plug that lets any agent use any service you already pay for Everyone is still paying for software, almost nobody is building the agents that make it free Save and watch it today Then read how to wire agents into your own stack in the harness engineering guide below ↓
3
3
1
77
22,122
This is f*cking insane. Palantir CEO Alex Karp: "There was like a holy f*ck, this is real. But somehow, it's not working. And we're not allowed to say it publicly, because we'll look stupid" In this 20-minute interview, the man who built Palantir with Peter Thiel says what most CEOs only say in private "If you run around saying AI allowed you to fire two-thirds of your workforce, you might as well just go sign up for the Bernie Sanders manifesto" "I've been telling them for six months, we're going to be nationalized. And they're like, why would anyone nationalize us? It never happened in America" Token Maxing → AI That Isn't Working → Mass Layoffs → Nationalization His fix isn't less AI. It's taste: knowing which problem is worth solving, then putting AI on it "If you upskill somebody, they're more valuable." The workers at the bottom who learned the tools got more valuable, not less Most companies are burning tokens to look busy, the people who win are the ones who know what to point AI at Bookmark it and watch it today Then read how to point agents at the problems that actually matter in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
7
5
1
66
19,115
This is f*cking insane. Chamath Palihapitiya, the billionaire who ran growth at early Facebook, just said software stopped being worth anything this week: "I think that IP and software was effectively rendered worthless two days ago When everything is headless, and you can wrap them in agent tools, and you can connect them via MCP to other experiences, the value of IP does no longer exist in software" He says people decompiled Adobe's 5 core products and published open-source versions. Moats that took decades, gone in days In 2 minutes, he explains why the software business just flipped Paid Software → Headless Services → Agent Tools → Software Worth $0 Headless means the app doesn't need a screen anymore, an agent calls it directly MCP is the plug that lets any agent use any service you already pay for Everyone is still paying for software, almost nobody is building the agents that make it free Save and watch it today Then read how to wire agents into your own stack in the harness engineering guide below ↓
27
mirrored retweeted
This is f*cking insane. Palantir CEO Alex Karp: "There was like a holy f*ck, this is real. But somehow, it's not working. And we're not allowed to say it publicly, because we'll look stupid" In this 20-minute interview, the man who built Palantir with Peter Thiel says what most CEOs only say in private "If you run around saying AI allowed you to fire two-thirds of your workforce, you might as well just go sign up for the Bernie Sanders manifesto" "I've been telling them for six months, we're going to be nationalized. And they're like, why would anyone nationalize us? It never happened in America" Token Maxing → AI That Isn't Working → Mass Layoffs → Nationalization His fix isn't less AI. It's taste: knowing which problem is worth solving, then putting AI on it "If you upskill somebody, they're more valuable." The workers at the bottom who learned the tools got more valuable, not less Most companies are burning tokens to look busy, the people who win are the ones who know what to point AI at Bookmark it and watch it today Then read how to point agents at the problems that actually matter in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
7
5
1
66
19,115
This is f*cking insane. Perplexity CEO Aravind Srinivas just named the biggest threat to the $5 trillion data center boom, and it's sitting on your desk: "The biggest threat to a data center is if the intelligence can be packed locally on a chip that's running on the device Imagine the AI watches the tasks you repeatedly do, adapts to you and over time starts automating a lot of the things you do. That's your intelligence, you own it, it's your brain" In just over 1 minute, he explains why the AI that takes over your busywork won't live in the cloud Cloud Chatbot → AI On Your Laptop → AI That Learns You → Your Own Second Brain Test time training means the model keeps learning after it ships, by watching what you actually do all day Local means none of it ever leaves your computer, so it can watch everything without a privacy nightmare Everyone is renting AI from a data center, almost nobody is building the agents that will run on their own machine Bookmark and watch it today Then read how to set up agents that automate your repeat work in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
7
26
5
227
36,630
This is f*cking insane. Palantir CEO Alex Karp: "There was like a holy f*ck, this is real. But somehow, it's not working. And we're not allowed to say it publicly, because we'll look stupid" In this 20-minute interview, the man who built Palantir with Peter Thiel says what most CEOs only say in private "If you run around saying AI allowed you to fire two-thirds of your workforce, you might as well just go sign up for the Bernie Sanders manifesto" "I've been telling them for six months, we're going to be nationalized. And they're like, why would anyone nationalize us? It never happened in America" Token Maxing → AI That Isn't Working → Mass Layoffs → Nationalization His fix isn't less AI. It's taste: knowing which problem is worth solving, then putting AI on it "If you upskill somebody, they're more valuable." The workers at the bottom who learned the tools got more valuable, not less Most companies are burning tokens to look busy, the people who win are the ones who know what to point AI at Bookmark it and watch it today Then read how to point agents at the problems that actually matter in the harness engineering guide below ↓
27
mirrored retweeted
This is f*cking insane. Perplexity CEO Aravind Srinivas just named the biggest threat to the $5 trillion data center boom, and it's sitting on your desk: "The biggest threat to a data center is if the intelligence can be packed locally on a chip that's running on the device Imagine the AI watches the tasks you repeatedly do, adapts to you and over time starts automating a lot of the things you do. That's your intelligence, you own it, it's your brain" In just over 1 minute, he explains why the AI that takes over your busywork won't live in the cloud Cloud Chatbot → AI On Your Laptop → AI That Learns You → Your Own Second Brain Test time training means the model keeps learning after it ships, by watching what you actually do all day Local means none of it ever leaves your computer, so it can watch everything without a privacy nightmare Everyone is renting AI from a data center, almost nobody is building the agents that will run on their own machine Bookmark and watch it today Then read how to set up agents that automate your repeat work in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
7
26
5
227
36,630
This is f*cking insane. Bill Gates just put a number on how long human labor has left: "My big concern is this 20-year transition period where there'll be winners and losers and a lot of disruption We will have robots that can build houses, grow food, do surgery, and none of the things that human labor provides to the economy will be necessary" In 85 seconds, the founder of Microsoft explains which side of the next 20 years you want to be standing on Chatbots → Agents → Humanoid Robots → Labor Nobody Needs → Winners And Losers Humanoid robots are 2-3 years away or 8, and he says nobody thinks it's unsolvable anymore China alone is pouring more capital and more companies into it than anyone can count And 20 years is his "safe" number. He says it could be less The winners won't be the people who work the hardest, they'll be the ones who already have AI agents doing the work for them Bookmark it and send it to someone who still thinks their job is safe Then read how to put your first AI agents to work in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
4
1
1
30
1,426
This is f*cking insane. Perplexity CEO Aravind Srinivas just named the biggest threat to the $5 trillion data center boom, and it's sitting on your desk: "The biggest threat to a data center is if the intelligence can be packed locally on a chip that's running on the device Imagine the AI watches the tasks you repeatedly do, adapts to you and over time starts automating a lot of the things you do. That's your intelligence, you own it, it's your brain" In just over 1 minute, he explains why the AI that takes over your busywork won't live in the cloud Cloud Chatbot → AI On Your Laptop → AI That Learns You → Your Own Second Brain Test time training means the model keeps learning after it ships, by watching what you actually do all day Local means none of it ever leaves your computer, so it can watch everything without a privacy nightmare Everyone is renting AI from a data center, almost nobody is building the agents that will run on their own machine Bookmark and watch it today Then read how to set up agents that automate your repeat work in the harness engineering guide below ↓
28
mirrored retweeted
This is f*cking insane. Bill Gates just put a number on how long human labor has left: "My big concern is this 20-year transition period where there'll be winners and losers and a lot of disruption We will have robots that can build houses, grow food, do surgery, and none of the things that human labor provides to the economy will be necessary" In 85 seconds, the founder of Microsoft explains which side of the next 20 years you want to be standing on Chatbots → Agents → Humanoid Robots → Labor Nobody Needs → Winners And Losers Humanoid robots are 2-3 years away or 8, and he says nobody thinks it's unsolvable anymore China alone is pouring more capital and more companies into it than anyone can count And 20 years is his "safe" number. He says it could be less The winners won't be the people who work the hardest, they'll be the ones who already have AI agents doing the work for them Bookmark it and send it to someone who still thinks their job is safe Then read how to put your first AI agents to work in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
4
1
1
30
1,426
This is f*cking insane. Microsoft founder Bill Gates says he's staking his whole reputation on this: "Something that exceeds human cognition, not just in one area like farming or mathematics, but across the board, and then in a few years coupled with low-cost, quite capable humanoid robots That is a more profound change to our economy and jobs than anything in my lifetime, by a dramatic amount" In 80 seconds, he explains why the PC and the internet created more jobs than they killed, and why with AI the past is only there to fool you Chatbots → Agents → AI Smarter Than Us At Everything → Humanoid Robots → A New Economy The man who put a computer on every desk is telling you this time is different The people who learn to put AI to work now will be on the right side of that change Bookmark it and send it to someone who still thinks AI is just another tech wave Then read how to put AI to work for you in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
48
45
6
276
68,473
Anothe Bill Gates talk ↓ xitter.cf/mirrorxbt/status/21194…
This is f*cking insane. Bill Gates just put a number on how long human labor has left: "My big concern is this 20-year transition period where there'll be winners and losers and a lot of disruption We will have robots that can build houses, grow food, do surgery, and none of the things that human labor provides to the economy will be necessary" In 85 seconds, the founder of Microsoft explains which side of the next 20 years you want to be standing on Chatbots → Agents → Humanoid Robots → Labor Nobody Needs → Winners And Losers Humanoid robots are 2-3 years away or 8, and he says nobody thinks it's unsolvable anymore China alone is pouring more capital and more companies into it than anyone can count And 20 years is his "safe" number. He says it could be less The winners won't be the people who work the hardest, they'll be the ones who already have AI agents doing the work for them Bookmark it and send it to someone who still thinks their job is safe Then read how to put your first AI agents to work in the harness engineering guide below ↓
1
90
This is f*cking insane. Perplexity CEO Aravind Srinivas on Joe Rogan: "Microsoft built this concept of a knowledge worker because they wanted to sell more office software. Putting a PC on every desk and getting you glued to it, that was Bill Gates If that part's going to be done by an AI, it's not necessarily a bad thing" In under 6 minutes, he explains why the desk job was a sales strategy, and what's left when AI takes it over Pen And Paper → A PC On Every Desk → Word And Excel → AI Doing The Desk Work Your office job was designed so someone could sell you the software for it The next one gets built around asking AI the right questions, not memorizing the answers School still grades kids on 20 answers any AI already has, the people who win will be the ones asking the questions Bookmark and watch it today Then read how to hand the desk work to agents in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
10
50
8
343
71,326
This is f*cking insane. Bill Gates just put a number on how long human labor has left: "My big concern is this 20-year transition period where there'll be winners and losers and a lot of disruption We will have robots that can build houses, grow food, do surgery, and none of the things that human labor provides to the economy will be necessary" In 85 seconds, the founder of Microsoft explains which side of the next 20 years you want to be standing on Chatbots → Agents → Humanoid Robots → Labor Nobody Needs → Winners And Losers Humanoid robots are 2-3 years away or 8, and he says nobody thinks it's unsolvable anymore China alone is pouring more capital and more companies into it than anyone can count And 20 years is his "safe" number. He says it could be less The winners won't be the people who work the hardest, they'll be the ones who already have AI agents doing the work for them Bookmark it and send it to someone who still thinks their job is safe Then read how to put your first AI agents to work in the harness engineering guide below ↓
218
mirrored retweeted
This is f*cking insane. Perplexity CEO Aravind Srinivas on Joe Rogan: "Microsoft built this concept of a knowledge worker because they wanted to sell more office software. Putting a PC on every desk and getting you glued to it, that was Bill Gates If that part's going to be done by an AI, it's not necessarily a bad thing" In under 6 minutes, he explains why the desk job was a sales strategy, and what's left when AI takes it over Pen And Paper → A PC On Every Desk → Word And Excel → AI Doing The Desk Work Your office job was designed so someone could sell you the software for it The next one gets built around asking AI the right questions, not memorizing the answers School still grades kids on 20 answers any AI already has, the people who win will be the ones asking the questions Bookmark and watch it today Then read how to hand the desk work to agents in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
10
50
8
343
71,326
perplexity ceo about sales strategy ↓ xitter.cf/mirrorxbt/status/21190…
This is f*cking insane. Perplexity CEO Aravind Srinivas on Joe Rogan: "Microsoft built this concept of a knowledge worker because they wanted to sell more office software. Putting a PC on every desk and getting you glued to it, that was Bill Gates If that part's going to be done by an AI, it's not necessarily a bad thing" In under 6 minutes, he explains why the desk job was a sales strategy, and what's left when AI takes it over Pen And Paper → A PC On Every Desk → Word And Excel → AI Doing The Desk Work Your office job was designed so someone could sell you the software for it The next one gets built around asking AI the right questions, not memorizing the answers School still grades kids on 20 answers any AI already has, the people who win will be the ones asking the questions Bookmark and watch it today Then read how to hand the desk work to agents in the harness engineering guide below ↓
173
mirrored retweeted
This is f*cking insane. Microsoft founder Bill Gates says he's staking his whole reputation on this: "Something that exceeds human cognition, not just in one area like farming or mathematics, but across the board, and then in a few years coupled with low-cost, quite capable humanoid robots That is a more profound change to our economy and jobs than anything in my lifetime, by a dramatic amount" In 80 seconds, he explains why the PC and the internet created more jobs than they killed, and why with AI the past is only there to fool you Chatbots → Agents → AI Smarter Than Us At Everything → Humanoid Robots → A New Economy The man who put a computer on every desk is telling you this time is different The people who learn to put AI to work now will be on the right side of that change Bookmark it and send it to someone who still thinks AI is just another tech wave Then read how to put AI to work for you in the harness engineering guide below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
48
45
6
276
68,473
This is f*cking insane. OpenAI CEO Sam Altman was asked why Ilya, Mira, Dario and Elon all left OpenAI to build their own labs. His answer: "There is something that happens in many people's brains where they only trust themselves to build this, because the stakes feel so high His approach is bad, and so I can't let power get concentrated in his hands" In 2 minutes, he also admits the hardest question on his desk: how smart an AI should be allowed to be compared to a person, and how fast it gets to improve itself Chatbots → Agents → Self-Improving AI → Who Gets To Control It The last time his heart was pounding was right before OpenAI announced its Navier-Stokes result, AI doing math that people could not The labs are fighting over who builds it. Your only move is learning to use it before that fight is over Bookmark it and watch it tonight Then read how to put agents to work for you in the enhanced edition of harness engineering below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
2
2
1
26
744
This is f*cking insane. Microsoft founder Bill Gates says he's staking his whole reputation on this: "Something that exceeds human cognition, not just in one area like farming or mathematics, but across the board, and then in a few years coupled with low-cost, quite capable humanoid robots That is a more profound change to our economy and jobs than anything in my lifetime, by a dramatic amount" In 80 seconds, he explains why the PC and the internet created more jobs than they killed, and why with AI the past is only there to fool you Chatbots → Agents → AI Smarter Than Us At Everything → Humanoid Robots → A New Economy The man who put a computer on every desk is telling you this time is different The people who learn to put AI to work now will be on the right side of that change Bookmark it and send it to someone who still thinks AI is just another tech wave Then read how to put AI to work for you in the harness engineering guide below ↓
46
mirrored retweeted
This is f*cking insane. OpenAI CEO Sam Altman was asked why Ilya, Mira, Dario and Elon all left OpenAI to build their own labs. His answer: "There is something that happens in many people's brains where they only trust themselves to build this, because the stakes feel so high His approach is bad, and so I can't let power get concentrated in his hands" In 2 minutes, he also admits the hardest question on his desk: how smart an AI should be allowed to be compared to a person, and how fast it gets to improve itself Chatbots → Agents → Self-Improving AI → Who Gets To Control It The last time his heart was pounding was right before OpenAI announced its Navier-Stokes result, AI doing math that people could not The labs are fighting over who builds it. Your only move is learning to use it before that fight is over Bookmark it and watch it tonight Then read how to put agents to work for you in the enhanced edition of harness engineering below ↓
Your agent passed every test The issue is still open A better prompt won't fix it, because the model isn't the broken part I broke down the harness, loop and graph around it: what each one owns, how they nest, and which one to fix first ↓
2
2
1
26
744
This is f*cking insane. Tesla and SpaceX founder Elon Musk just said he'd put serious money on this: "10 years from now, there will be well over a billion humanoid robots, and the productivity per robot will be probably five times that of a human Meaning the billion humanoid robots will be more productive than all humans combined, and I think this is a conservative estimate" In 44 seconds, he explains why the next 10 years decide who owns the robots and who competes with them Chatbots → Agents → AI Workers → Humanoid Robots → An Economy Bigger Than All Humans Combined Getting ready for this doesn't mean watching robot videos It means learning to put AI to work now, while the robots are still learning to walk 95% will end up competing with machines that never sleep, the 5% who run them will own what comes next Worth more than any $5,000 robotics course Bookmark it and watch it tonight Then read how to put your first AI workers to work in the harness engineering guide below ↓
3
4
1
29
885
This is f*cking insane. OpenAI CEO Sam Altman was asked why Ilya, Mira, Dario and Elon all left OpenAI to build their own labs. His answer: "There is something that happens in many people's brains where they only trust themselves to build this, because the stakes feel so high His approach is bad, and so I can't let power get concentrated in his hands" In 2 minutes, he also admits the hardest question on his desk: how smart an AI should be allowed to be compared to a person, and how fast it gets to improve itself Chatbots → Agents → Self-Improving AI → Who Gets To Control It The last time his heart was pounding was right before OpenAI announced its Navier-Stokes result, AI doing math that people could not The labs are fighting over who builds it. Your only move is learning to use it before that fight is over Bookmark it and watch it tonight Then read how to put agents to work for you in the enhanced edition of harness engineering below ↓
35