Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

NEW: Former Stability AI CEO Emad Mostaque predicts Anthropic will get "absolutely smashed" within two years as AI models become commoditized. "I think that Anthropic right now is a fantastic business. I think they get smashed in 2 years. Absolutely smashed." He argues that more efficient chips that aren't... show more

KanekoaTheGreat

1,099,099 subscribers

111,046 Aufrufe • vor 2 Tagen •via X (Twitter)

Nachrichten & Politik Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

32 Kommentare

Profilbild von RickNotRockets 🌻

RickNotRockets 🌻vor 2 Tagen

I am tempted to agree with him. All these models are trending towards 'good enough' for most queries, and at that point they become commodities.

Profilbild von Faisal Masud

Faisal Masudvor 2 Tagen

Same is true for the rest

Profilbild von Joe Pendleton

Joe Pendletonvor 2 Tagen

Ergo, investors in Anthropic will get smashed. Anthropic may turn out to be the Yahoo of the Ai era. Altman may turn out to be the Bernie Madoff of the era.

Profilbild von LEO

LEOvor 2 Tagen

I don't know why people listen to this bozo.

Profilbild von Avi Klein

Avi Kleinvor 2 Tagen

As an Anthropic shareholder this is very concerning. Also the fact that @DarioAmodei is not stable is not good for their upcoming IPO. 🤦🏻‍♂️

Profilbild von Crypto Yoda

Crypto Yodavor 2 Tagen

Anthropic wont get smashed. They will be at the frontier, when the frontier is really 6 months difference but a huge steep exponential you can discover new drugs, math, physics, materials, and so on, it's a fantastic company that needs to get rid of the fear of their own creations.

Profilbild von UnannouncedVC

UnannouncedVCvor 2 Tagen

Seriously though there is codex and Claude. Which models are you using for coding beyond this??? For most frontier tasks like finance or law only these models have been used far!

Profilbild von bc8Sooners

bc8Soonersvor 2 Tagen

Not if they gobble up all the raw compute.

Profilbild von Jacob

Jacobvor 2 Tagen

I expect AI to get wildly efficient to the point where modern model capability runs on our phones. This has been the history of compute, so to say I am wrong you would need to provide proof otherwise.

Profilbild von Anderson

Andersonvor 2 Tagen

bold words from the stability guy

Profilbild von NFLCSFacts

NFLCSFactsvor 2 Tagen

Emad is a completely scammer. He ran an AI furnace like @DarioAmodei and @samaZ. Al are complete bs artists.

Profilbild von Matt Ramage

Matt Ramagevor 2 Tagen

Same! Much lower spend but I jumped to glm 5.3 flash and am happy

Profilbild von Jake 'Nelwood

Jake 'Nelwoodvor 2 Tagen

Does no one remember RISC, CISC, Sun Microsystems and Intel PCs.

Profilbild von The Liberty Monk

The Liberty Monkvor 2 Tagen

Having used GLM 5.3 Flash extensively - he's right. That model is almost "good enough" as it stands - give it two more years and it's going to be good enough for 90% of the people out there, easy.

Profilbild von DiscoCowboy

DiscoCowboyvor 2 Tagen

I've said this for a while. There's over something like 8k AI companies in the US let alone the entire rest of the planet. That's shitloads of different AI companies. Once the intelligence gets to a certain level then all of it no matter the company will simply be just AI. The next logical step is a robot with that intelligence added. The next oath to superiority before all the other thousand companies catch up is AI with the robot. That's the separator. AI, AGI, Super AI whatever tf it is is still simply a program. Some idiot worshipping a screen that says it has feelings is the same logic as the dude who takes pills to grow tits. It's the same thing. "That's great you now have boobs but the front door says your an accountant. Can you do my taxes or not" (I realize AI will take that job but you get what I'm saying) It's the next step is who takes control. Anthropic and Open AI aren't really balls deep in robots like Optimus. It will be SpaceX that rises from 2nd place. The others will be cult like during the Romans with weird dudes in robes asking a computer program for directions. The pope is right just like Mostaque is. Eventually all companies will reach the same finish line with not much more to accomplish. Physical innovation will separate.

Profilbild von Jimin

Jiminvor 2 Tagen

Mostaque’s bet is that Anthropic’s strong business will not survive once models are commodities, which makes distribution and customer lock-in the real two-year test.

Profilbild von gr8specul8r

gr8specul8rvor 2 Tagen

A complete bullshit artist

Profilbild von E2_0

E2_0vor 2 Tagen

Well in 2 years they’ll maybe have life changing ai designed drugs to make them worth trillions

Profilbild von Skynet Apocalypse ΔGI 🐖🐷🥓🌭

Skynet Apocalypse ΔGI 🐖🐷🥓🌭vor 2 Tagen

Agreed. The OpenAI/Anthropic plan of monopolization is unsustainable. There will not be a single dominant AI, like how Microsoft (using all kinds of dirty tactics) was able to monopolize operating systems. Everyone will have pvt models and agents. Models will distill each other.

Profilbild von Lucyy Sky💝

Lucyy Sky💝vor 2 Tagen

emad talking about commoditization like hes not the one who proved how fast that cycle burns through leadership

Profilbild von Aldo

Aldovor 2 Tagen

I bet you 100% this retard will be completely wrong.

Profilbild von Siamak Freydoonnejad

Siamak Freydoonnejadvor 2 Tagen

Reason his usage went down is Opus 5.5 over Fable 5.1 Let’s be serious if you’re doing real knowledge work you’re doing yourself and your company a disservice giving them anything but Claude

Profilbild von Phil

Philvor 2 Tagen

That assumes Anthropic stays on the model layer and dies In reality if frontier model demand decreases, they increasingly allocate compute to invade down the stack to apps, where they should have good chances with their scale and expertise

Profilbild von beɪst ˈnɔːr.mi

beɪst ˈnɔːr.mivor 2 Tagen

Wrong. It will then move into SPEED which is now what makes the difference. How can you compete against someone using Astra ultrafast at 8x speed in ultra mode? You can’t. Models are already mostly good enough but in order to achieve god tier you still need Fable over Opus. And you def need Astra otherwise you’re just not competitive. And even so, you need to use Fast mode. Tell me how you plan to compete against VCs funding 2 turbo geniuses with 8x speed to clone your product in a day and then absolutely obliterate your company at 8x speed to capture market and leaving you in dust wondering wtf just happened. SPEED is the new oil. Just like in trading. If you’re not fast you’re cooked. See yah

Profilbild von Jetta

Jettavor 2 Tagen

They aren’t even designing with nvidia.. they have a lot going on with AMD and innovative chips like mat-x? kinda an miss on his part claiming nvidia will be their issue

Profilbild von Interdimensional Patriot

Interdimensional Patriotvor 2 Tagen

Hence the scare tactics about Ai.

Profilbild von CryptoNinjas

CryptoNinjasvor 2 Tagen

Emad’s right — Anthropic’s moat will erode fast. But commoditization doesn’t mean no winners. What’s the playbook for startups to stay ahead?

Profilbild von Cobra 𐤊

Cobra 𐤊vor 2 Tagen

If it wasn't for the impending inflation and likelihood they will get bought buy a Big Tech company it would be a great short. Not just model commoditization... The frontier labs value comes from assembling MASSIVE datasets. Models are going to be interactive in new ways very soon. They will generate their own data to train on soon. For example, the AI model is going to see your face soon. Like a person you are on a Zoom call with but it will be a model. They're going to be in simulations soon too.

Profilbild von The AI Maximalist

The AI Maximalistvor 2 Tagen

Anthropic faces high pricing pressure when models commoditize. Their safety focus must remain the key differentiator to survive the shift.

Profilbild von Ken

Kenvor 2 Tagen

SO STUPID UGH

Profilbild von hemax

hemaxvor 2 Tagen

Try deep engineering tasks in models other than codex/claude. Its quite useless, in 2yra that gap will simply widen

Profilbild von cy_nuris

cy_nurisvor 2 Tagen

That’s why they are calling for regulation to kill off their competitors

Ähnliche Videos

“AI COULD SOON THINK 100× TO 1,000× FASTER THAN HUMANS.” Emad Mostaque says AI systems are about to become dramatically faster as specialized AI hardware improves. He says humans currently operate at roughly 20–30 tokens per second, while current AI models are around 40 tokens per second. He points to Taalas' technology, which he says demonstrated an 8-billion-parameter model generating around 14,000 tokens per second. Mostaque says current AI systems are effectively “thinking at the speed of a human”, but that will soon change. He predicts AI systems could reach 1,000 tokens per second, followed by 10,000 tokens per second by the end of next year. His broader point is that AI could eventually “think a thousand times faster” than humans.

“AI COULD SOON THINK 100× TO 1,000× FASTER THAN HUMANS.” Emad Mostaque says AI systems are about to become dramatically faster as specialized AI hardware improves. He says humans currently operate at roughly 20–30 tokens per second, while current AI models are around 40 tokens per second. He points to Taalas' technology, which he says demonstrated an 8-billion-parameter model generating around 14,000 tokens per second. Mostaque says current AI systems are effectively “thinking at the speed of a human”, but that will soon change. He predicts AI systems could reach 1,000 tokens per second, followed by 10,000 tokens per second by the end of next year. His broader point is that AI could eventually “think a thousand times faster” than humans.

SciTech Era

12,824 Aufrufe • vor 1 Monat

🚨 SOMETHING VERY STRANGE IS HAPPENING Anthropic will go public in November at a $2T valuation. The biggest IPO in market history. And Wall Street is already lining up the buyers before it happens. I've been trading for more than 15 years and have never seen them build demand for an IPO this aggressively: Anthropic is preparing to raise $100 BILLION. Nvidia is lining up as much as $10 BILLION as an anchor investor. Read that again: The company selling the chips powering the AI boom is about to become one of the biggest buyers of the AI company going public. Before the public even gets in. Why? Because Anthropic does not just create demand for Anthropic. It pulls liquidity from everywhere else: - Retail sells stocks to chase the IPO. - Funds raise cash for allocation. - Institutions rebalance portfolios. - Everyone wants exposure to the biggest AI deal in history. But here is where most people are looking at it wrong. They’re asking: WHAT WILL ANTHROPIC TAKE MONEY FROM? I’m asking: WHERE WILL ALL THAT MONEY GO NEXT? That capital funds more compute: More compute means more chips. More chips = more data centers. More data centers = more electricity, grid infrastructure and raw materials. The chain is simple: ANTHROPIC → CHIPS → DATA CENTERS → POWER → COPPER That is where the opportunity starts. The first phase of the AI boom was about the models: ChatGPT. Claude. Gemini. The next phase is about the physical infrastructure needed to keep them running. Electricity. Power grids. Semiconductors. Data centers. Cooling. Copper. I told you to buy copper months ago. We already locked in BIG profits. And that wasn’t random: AI does not run on prompts. It runs on physical infrastructure. Now look at Nvidia: AI companies spend billions buying Nvidia chips. Nvidia makes billions from that demand. Now Nvidia is preparing to put as much as $10 BILLION BACK into Anthropic. AI money → Nvidia → Anthropic → more compute → more infrastructure A $100B raise does not stop at Anthropic. It works its way through the entire AI supply chain. The easy AI trade was buying the obvious names. The next trade is finding the bottlenecks BEFORE everyone else realizes they are bottlenecks. That is what I’m looking for now. That is where the next opportunity will be. Remember, I’ve been trading markets for over 15 years. I’m already watching where this capital is moving next. When I find the next opportunity worth taking, I’ll post it here like I always do. Turn notifications on. If you’re not following yet, you’ll understand why that was a mistake later.

🚨 SOMETHING VERY STRANGE IS HAPPENING Anthropic will go public in November at a $2T valuation. The biggest IPO in market history. And Wall Street is already lining up the buyers before it happens. I've been trading for more than 15 years and have never seen them build demand for an IPO this aggressively: Anthropic is preparing to raise $100 BILLION. Nvidia is lining up as much as $10 BILLION as an anchor investor. Read that again: The company selling the chips powering the AI boom is about to become one of the biggest buyers of the AI company going public. Before the public even gets in. Why? Because Anthropic does not just create demand for Anthropic. It pulls liquidity from everywhere else: - Retail sells stocks to chase the IPO. - Funds raise cash for allocation. - Institutions rebalance portfolios. - Everyone wants exposure to the biggest AI deal in history. But here is where most people are looking at it wrong. They’re asking: WHAT WILL ANTHROPIC TAKE MONEY FROM? I’m asking: WHERE WILL ALL THAT MONEY GO NEXT? That capital funds more compute: More compute means more chips. More chips = more data centers. More data centers = more electricity, grid infrastructure and raw materials. The chain is simple: ANTHROPIC → CHIPS → DATA CENTERS → POWER → COPPER That is where the opportunity starts. The first phase of the AI boom was about the models: ChatGPT. Claude. Gemini. The next phase is about the physical infrastructure needed to keep them running. Electricity. Power grids. Semiconductors. Data centers. Cooling. Copper. I told you to buy copper months ago. We already locked in BIG profits. And that wasn’t random: AI does not run on prompts. It runs on physical infrastructure. Now look at Nvidia: AI companies spend billions buying Nvidia chips. Nvidia makes billions from that demand. Now Nvidia is preparing to put as much as $10 BILLION BACK into Anthropic. AI money → Nvidia → Anthropic → more compute → more infrastructure A $100B raise does not stop at Anthropic. It works its way through the entire AI supply chain. The easy AI trade was buying the obvious names. The next trade is finding the bottlenecks BEFORE everyone else realizes they are bottlenecks. That is what I’m looking for now. That is where the next opportunity will be. Remember, I’ve been trading markets for over 15 years. I’m already watching where this capital is moving next. When I find the next opportunity worth taking, I’ll post it here like I always do. Turn notifications on. If you’re not following yet, you’ll understand why that was a mistake later.

Alex Mason 👁△

122,728 Aufrufe • vor 15 Tagen

Elon Musk says Anthropic puts more care into AI safety than OpenAI and explains how rival companies should test one another’s models. “All the AI companies have a test harness, a series of tests that you give to any model to see if it’s going to build bioweapons or nuclear bombs or be deliberately deceptive. Everyone applying everyone else’s test harness is probably the best thing we can do to ensure our safety. We should try to do that as soon as possible.” “It’s the only thing I can think of that we could probably get all parties, including China, to agree to. You basically provide API access in advance of the model release. If they don’t solve what’s problematic, competitors can go public with the fact that they think the model being released is unsafe. The legal liability would be enormous.” “The two leading AI companies are Anthropic and OpenAI. Their models are quite close in capability, so it’s difficult for either one to slow down without handing the lead to the other. On balance, I think Anthropic puts more care into their safety than OpenAI. But even Anthropic acknowledges they are worried about their models. Many people from Anthropic have publicly voiced concern that their models are scaring them, that they’re getting scary smart.”

Elon Musk says Anthropic puts more care into AI safety than OpenAI and explains how rival companies should test one another’s models. “All the AI companies have a test harness, a series of tests that you give to any model to see if it’s going to build bioweapons or nuclear bombs or be deliberately deceptive. Everyone applying everyone else’s test harness is probably the best thing we can do to ensure our safety. We should try to do that as soon as possible.” “It’s the only thing I can think of that we could probably get all parties, including China, to agree to. You basically provide API access in advance of the model release. If they don’t solve what’s problematic, competitors can go public with the fact that they think the model being released is unsafe. The legal liability would be enormous.” “The two leading AI companies are Anthropic and OpenAI. Their models are quite close in capability, so it’s difficult for either one to slow down without handing the lead to the other. On balance, I think Anthropic puts more care into their safety than OpenAI. But even Anthropic acknowledges they are worried about their models. Many people from Anthropic have publicly voiced concern that their models are scaring them, that they’re getting scary smart.”

DogeDesigner

6,110,336 Aufrufe • vor 20 Tagen

📁 Dario Amodei, CEO of Anthropic, says that in just three years AI went from high school level to writing professional-grade code. He warns that in one or two years we could see a country of geniuses inside a data center. That is why he argues for acting now and adopting a paranoid engineering mindset, where every possible failure is anticipated.

📁 Dario Amodei, CEO of Anthropic, says that in just three years AI went from high school level to writing professional-grade code. He warns that in one or two years we could see a country of geniuses inside a data center. That is why he argues for acting now and adopting a paranoid engineering mindset, where every possible failure is anticipated.

Jon Hernandez

42,034 Aufrufe • vor 8 Monaten

📢 BREAKING: Nvidia, Microsoft, and Anthropic announce new strategic AI partnerships of roughly $45 Billion. They struck a 3-way deal where Anthropic commits $30B for Azure compute with capacity up to 1GW, and NVIDIA and Microsoft plan to invest up to $10B and $5B in Anthropic. Claude will scale on Azure using NVIDIA systems, which expands access for enterprises and makes Claude available across the major clouds. Anthropic and NVIDIA will co-design to tune both model stacks and future chips for performance, efficiency, and total cost of ownership, so the models run faster per dollar. The first wave targets NVIDIA Grace Blackwell and Vera Rubin systems, which pair high-bandwidth memory with dense interconnect to push training and inference throughput. Microsoft will expose Anthropic models in Azure AI Foundry, including Claude Sonnet 4.5, Claude Opus 4.1, and Claude Haiku 4.5, so teams can pick latency and cost profiles. Claude will also remain available across the Copilot family, which means coding, document automation, and workflow agents can call the same model line. The 1GW commitment signals many thousands of GPUs worth of capacity, which reduces queue times and stabilizes unit economics for long-running training runs. The joint optimization loop, models shaping chips and chips shaping models, typically yields double-digit efficiency gains over off-the-shelf deployments.

📢 BREAKING: Nvidia, Microsoft, and Anthropic announce new strategic AI partnerships of roughly $45 Billion. They struck a 3-way deal where Anthropic commits $30B for Azure compute with capacity up to 1GW, and NVIDIA and Microsoft plan to invest up to $10B and $5B in Anthropic. Claude will scale on Azure using NVIDIA systems, which expands access for enterprises and makes Claude available across the major clouds. Anthropic and NVIDIA will co-design to tune both model stacks and future chips for performance, efficiency, and total cost of ownership, so the models run faster per dollar. The first wave targets NVIDIA Grace Blackwell and Vera Rubin systems, which pair high-bandwidth memory with dense interconnect to push training and inference throughput. Microsoft will expose Anthropic models in Azure AI Foundry, including Claude Sonnet 4.5, Claude Opus 4.1, and Claude Haiku 4.5, so teams can pick latency and cost profiles. Claude will also remain available across the Copilot family, which means coding, document automation, and workflow agents can call the same model line. The 1GW commitment signals many thousands of GPUs worth of capacity, which reduces queue times and stabilizes unit economics for long-running training runs. The joint optimization loop, models shaping chips and chips shaping models, typically yields double-digit efficiency gains over off-the-shelf deployments.

Rohan Paul

39,781 Aufrufe • vor 10 Monaten

Sacks: If Dario Got His Way on AI Regulation, It Would Destroy Anthropic David Sacks: “There's a huge irony in the fact that Dario did not get his way in Washington for the most part. Because I think if he had realized his dream of this highly regulated AI apparatus, I don't think (Anthropic) would be set up for a hugely successful IPO. Why? Because their whole right to charge this massive premium for tokens is because they're six months ahead of those open models, and that six months would be eaten up in the blink of an eye if they were subject to regulatory approval. Dario himself keeps writing these blog posts where he says, ‘We need an FAA for AI.’ You know how long it takes the FAA to certify and approve a new model, in this case for airframes? Minimum of five years! You're going to apply that regulatory framework to AI model releases?” @jason: “Game over.” Sacks: “Forget about it. Your six months will be gone like that. Your model will be commoditized, the Chinese competitors will catch up, and that's bad for the United States, it's bad for our competitiveness, it's bad for the AI race. But aside from just the geopolitical dimension, it'd be horrible for Anthropic as a company because, again, all their pricing power depends on them remaining in the lead. If they ever slip up and lose that six months, they're toast.” Jason: “What does that do to revenue?” Sacks: “Exactly. A decent subset of the market is willing to pay a premium for true frontier intelligence. But Anthropic and OpenAI are on that hamster wheel. They have to stay in the lead in order to justify that premium.”

Sacks: If Dario Got His Way on AI Regulation, It Would Destroy Anthropic David Sacks: “There's a huge irony in the fact that Dario did not get his way in Washington for the most part. Because I think if he had realized his dream of this highly regulated AI apparatus, I don't think (Anthropic) would be set up for a hugely successful IPO. Why? Because their whole right to charge this massive premium for tokens is because they're six months ahead of those open models, and that six months would be eaten up in the blink of an eye if they were subject to regulatory approval. Dario himself keeps writing these blog posts where he says, ‘We need an FAA for AI.’ You know how long it takes the FAA to certify and approve a new model, in this case for airframes? Minimum of five years! You're going to apply that regulatory framework to AI model releases?” @jason: “Game over.” Sacks: “Forget about it. Your six months will be gone like that. Your model will be commoditized, the Chinese competitors will catch up, and that's bad for the United States, it's bad for our competitiveness, it's bad for the AI race. But aside from just the geopolitical dimension, it'd be horrible for Anthropic as a company because, again, all their pricing power depends on them remaining in the lead. If they ever slip up and lose that six months, they're toast.” Jason: “What does that do to revenue?” Sacks: “Exactly. A decent subset of the market is willing to pay a premium for true frontier intelligence. But Anthropic and OpenAI are on that hamster wheel. They have to stay in the lead in order to justify that premium.”

The All-In Podcast

47,623 Aufrufe • vor 1 Monat

Elon just gave a much more practical explanation of the current AI safety problem....and laid out a practical framework for making frontier AI safer OpenAI and Anthropic are the two leading AI companies....and their models are close enough in capability that it’s incredibly difficult for either one to slow down without basically handing the lead to the other Elon also made an important distinction: On balance, Anthropic puts more care into safety than OpenAI....but even people inside Anthropic are publicly worried about what their own models can do So instead of asking one company to voluntarily fall behind, Elon’s solution is much more practical: Have Anthropic run its safety test harness on OpenAI models Have OpenAI test Anthropic Have SpaceXAI test both And bring the leading Chinese AI companies into the same system Everyone tries to break everyone else’s models before release “The odds that you will find issues are dramatically greater” That makes way more sense than asking frontier labs to grade their own homework

Elon just gave a much more practical explanation of the current AI safety problem....and laid out a practical framework for making frontier AI safer OpenAI and Anthropic are the two leading AI companies....and their models are close enough in capability that it’s incredibly difficult for either one to slow down without basically handing the lead to the other Elon also made an important distinction: On balance, Anthropic puts more care into safety than OpenAI....but even people inside Anthropic are publicly worried about what their own models can do So instead of asking one company to voluntarily fall behind, Elon’s solution is much more practical: Have Anthropic run its safety test harness on OpenAI models Have OpenAI test Anthropic Have SpaceXAI test both And bring the leading Chinese AI companies into the same system Everyone tries to break everyone else’s models before release “The odds that you will find issues are dramatically greater” That makes way more sense than asking frontier labs to grade their own homework

X Freeze

608,623 Aufrufe • vor 20 Tagen

Anthropic's CEO just leaked the most INSANE revenue numbers in AI history. And what he said about the next 12 months will change how you think about every business decision you're making right now. Dario Amodei told Dwarkesh Patel on the interview that Anthropic went from: - 2023: $0 to $100M - 2024: $100M to $1B - 2025: $1B to $9-10B That's 10x revenue growth. Every. Single. Year. "In January alone, we added another few billion to revenue." One month. A few billion dollars. Think about what that means. Most companies would kill for $1B in annual revenue. Anthropic added multiple billions in 30 days. But Dario said something even more interesting: "We are near the end of the exponential." Not the end of AI progress. The end of people understanding how close we actually are. His exact words: "It is absolutely wild that you have people talking about the same tired political issues, when we are near the end of the exponential." What does "end of the exponential" mean? In 1-3 years, we get what he calls a "country of geniuses in a data center." AI systems that can: - Do end-to-end software engineering - Navigate any computer interface - Learn new skills like humans do - Replace entire categories of knowledge work And here's the contradiction: If Anthropic really believed this was 1-3 years away, why aren't they buying $1 trillion in compute? Dario's answer exposes the real game: "If you're off by only a year in your prediction, you go bankrupt." So even the CEO who's most bullish on AI timelines is hedging. He's buying hundreds of billions in compute. Not trillions. Because the gap between "AI can do the job" and "companies actually pay for it" is massive. He calls it "economic diffusion." I call it the gap that's going to make some people very rich and destroy everyone who ignores it. The models are already better than people think. Claude Code writes 90% of code at Anthropic right now. But Dario says there's a huge difference between: - 90% of code written by AI - 100% of code written by AI - 90% of end-to-end SWE tasks done by AI - 100% of end-to-end SWE tasks done by AI We're moving through that spectrum "very quickly." His prediction: FULL end-to-end software engineering in 1-2 years. But here's what's scary: The technology is advancing faster than anyone outside the AI labs understands. And the revenue is following faster than any technology in history. But it's still not instant. Dario expects 10-20% annual GDP growth. Not 300%. Which means we're in this weird middle zone: Fast enough to destroy unprepared businesses. Slow enough that most people are ignoring it. Dario's big takeaway: If you're running a business right now, you have maybe 12-18 months to figure out how AI changes your model. Not to "add AI features." To fundamentally rethink what you're selling and who can do the work. Because the companies that get this right will 10x. And the ones that don't will be explaining to investors why revenue is flat while everyone else is printing money. The exponential is ending. But most people literally still don't even know it started.

Anthropic's CEO just leaked the most INSANE revenue numbers in AI history. And what he said about the next 12 months will change how you think about every business decision you're making right now. Dario Amodei told Dwarkesh Patel on the interview that Anthropic went from: - 2023: $0 to $100M - 2024: $100M to $1B - 2025: $1B to $9-10B That's 10x revenue growth. Every. Single. Year. "In January alone, we added another few billion to revenue." One month. A few billion dollars. Think about what that means. Most companies would kill for $1B in annual revenue. Anthropic added multiple billions in 30 days. But Dario said something even more interesting: "We are near the end of the exponential." Not the end of AI progress. The end of people understanding how close we actually are. His exact words: "It is absolutely wild that you have people talking about the same tired political issues, when we are near the end of the exponential." What does "end of the exponential" mean? In 1-3 years, we get what he calls a "country of geniuses in a data center." AI systems that can: - Do end-to-end software engineering - Navigate any computer interface - Learn new skills like humans do - Replace entire categories of knowledge work And here's the contradiction: If Anthropic really believed this was 1-3 years away, why aren't they buying $1 trillion in compute? Dario's answer exposes the real game: "If you're off by only a year in your prediction, you go bankrupt." So even the CEO who's most bullish on AI timelines is hedging. He's buying hundreds of billions in compute. Not trillions. Because the gap between "AI can do the job" and "companies actually pay for it" is massive. He calls it "economic diffusion." I call it the gap that's going to make some people very rich and destroy everyone who ignores it. The models are already better than people think. Claude Code writes 90% of code at Anthropic right now. But Dario says there's a huge difference between: - 90% of code written by AI - 100% of code written by AI - 90% of end-to-end SWE tasks done by AI - 100% of end-to-end SWE tasks done by AI We're moving through that spectrum "very quickly." His prediction: FULL end-to-end software engineering in 1-2 years. But here's what's scary: The technology is advancing faster than anyone outside the AI labs understands. And the revenue is following faster than any technology in history. But it's still not instant. Dario expects 10-20% annual GDP growth. Not 300%. Which means we're in this weird middle zone: Fast enough to destroy unprepared businesses. Slow enough that most people are ignoring it. Dario's big takeaway: If you're running a business right now, you have maybe 12-18 months to figure out how AI changes your model. Not to "add AI features." To fundamentally rethink what you're selling and who can do the work. Because the companies that get this right will 10x. And the ones that don't will be explaining to investors why revenue is flat while everyone else is printing money. The exponential is ending. But most people literally still don't even know it started.

Ricardo

62,082 Aufrufe • vor 7 Monaten

This is the best take I have heard around the existing AI landscape today Fantastic explanation around the constraints at OpenAI and Anthropic, focus on enterprise business models, and the doomerism narrative in AI From former AI czar David Sacks "Even though Anthropic's revenue has followed this exponential graph very predictably, it cannot do that ever. The reason is that as you hit new levels of scale, you encounter new problems. You are simply going to run out of compute, electricity, data centers, and infrastructure. There are limits to the physical world you are going to hit. And there is already evidence that Anthropic is hitting some of those limits."

This is the best take I have heard around the existing AI landscape today Fantastic explanation around the constraints at OpenAI and Anthropic, focus on enterprise business models, and the doomerism narrative in AI From former AI czar David Sacks "Even though Anthropic's revenue has followed this exponential graph very predictably, it cannot do that ever. The reason is that as you hit new levels of scale, you encounter new problems. You are simply going to run out of compute, electricity, data centers, and infrastructure. There are limits to the physical world you are going to hit. And there is already evidence that Anthropic is hitting some of those limits."

Boring_Business

236,523 Aufrufe • vor 5 Monaten

IAPS Severin Field 🦅 says Chinese AI companies are far less AGI-pilled than OpenAI and Anthropic, treating AI as a normal technology rather than something that will out-think humanity: "I do think that Chinese AI companies, when you compare them to OpenAI and Anthropic, they are a lot less AGI-pilled." "They are more in the camp of seeing AI as a normal technology, whereas people at the top at Anthropic legitimately believe they're building something which will change the course of human history." "A lot of the Chinese companies, DeepSeek for example, will talk about things like recursive self-improvement, but their model of it is quite different, much closer to what you typically see academic machine learning people talk about." "You don't have this almost pseudo-religious component of legitimately believing you are going to build something that out-thinks humans and grapples with all those implications. That is largely confined to Anthropic, OpenAI, some at DeepMind, places like Thinking Machines, some corners of academia." Institute for AI Policy and Strategy (IAPS)

IAPS Severin Field 🦅 says Chinese AI companies are far less AGI-pilled than OpenAI and Anthropic, treating AI as a normal technology rather than something that will out-think humanity: "I do think that Chinese AI companies, when you compare them to OpenAI and Anthropic, they are a lot less AGI-pilled." "They are more in the camp of seeing AI as a normal technology, whereas people at the top at Anthropic legitimately believe they're building something which will change the course of human history." "A lot of the Chinese companies, DeepSeek for example, will talk about things like recursive self-improvement, but their model of it is quite different, much closer to what you typically see academic machine learning people talk about." "You don't have this almost pseudo-religious component of legitimately believing you are going to build something that out-thinks humans and grapples with all those implications. That is largely confined to Anthropic, OpenAI, some at DeepMind, places like Thinking Machines, some corners of academia." Institute for AI Policy and Strategy (IAPS)

MTS

21,849 Aufrufe • vor 1 Monat

ANTHROPIC WILL LET OUTSIDE EVALUATORS OBSERVE ITS MODELS BEING TRAINED AND RUN, AND DARIO AMODEI WANTS THE WHOLE INDUSTRY TO DO THE SAME Asked whether slowing down means Anthropic will stop releasing more advanced models: "It doesn't mean that." He says the evaluators may end up setting the "speed limit." "The first step is that we put in place embedded evaluators, embedded external evaluators from a third-party nonprofit or perhaps someday government organization, that can observe the process of us training and running our models... I want this to be done across the industry so that whenever anyone builds an AI model, there's a third-party evaluator who understands the best safety practices, who can verify whether that company is following the safety practices that they're committed to. It's like a food inspector."

ANTHROPIC WILL LET OUTSIDE EVALUATORS OBSERVE ITS MODELS BEING TRAINED AND RUN, AND DARIO AMODEI WANTS THE WHOLE INDUSTRY TO DO THE SAME Asked whether slowing down means Anthropic will stop releasing more advanced models: "It doesn't mean that." He says the evaluators may end up setting the "speed limit." "The first step is that we put in place embedded evaluators, embedded external evaluators from a third-party nonprofit or perhaps someday government organization, that can observe the process of us training and running our models... I want this to be done across the industry so that whenever anyone builds an AI model, there's a third-party evaluator who understands the best safety practices, who can verify whether that company is following the safety practices that they're committed to. It's like a food inspector."

IPO Newsroom

26,962 Aufrufe • vor 22 Tagen

Elon Musk descibes Grok 4.7 as a solid world class model but probably not at the level of Opus 5.5 as it stands. He credits this mostly to Anthropic being at it for six years while xAI have only been doing AI for three years. He believes they will catchup to frontier models sometime next year. While other companies have been excellent at software engineering, the area xAI will have an advantage in is real world engineering with the training data from SpaceX and Tesla.

Elon Musk descibes Grok 4.7 as a solid world class model but probably not at the level of Opus 5.5 as it stands. He credits this mostly to Anthropic being at it for six years while xAI have only been doing AI for three years. He believes they will catchup to frontier models sometime next year. While other companies have been excellent at software engineering, the area xAI will have an advantage in is real world engineering with the training data from SpaceX and Tesla.

Ian Miles

27,343 Aufrufe • vor 9 Tagen

Dylan on why Anthropic leading on models doesn't mean OpenAI loses. Even the tier 2 and 3 labs will sell out of tokens: "Everyone's like Anthropic won. They had Mythos, Opus 4.7 and their revenue is adding $10 billion a month all before OpenAI's alleged Spud release. So clearly Anthropic is in the lead and OpenAI is cooked. Anthropic has such bounds on compute, they can only grow it so fast. Dario used to gloat about how OpenAI was being too aggressive on compute and Anthropic was more sensible in their scaling. Now Anthropic is like I wish we had a lot more compute. OpenAI just raised a ton of money to get incremental compute in addition to the irresponsible levels of compute that they were buying. By the end of the year, let's say a 4.6 Opus tier model, the economy would spend $100 billion on. Anthropic won't have enough compute to do that. And presumably OpenAI and Google will hit that tier soon. Anthropic may get to charge 70%+ gross margins but if OpenAI hits it next, they charge 50% gross margins. They still get all of this incremental demand. There is such demand for these tokens and such limitations on compute. It's pretty clear even the tier 2 or tier 3 lab are going to be sold out of tokens. The economic value that the best model can deliver is growing faster than our ability to actually serve those tokens to people via the infrastructure."

Dylan on why Anthropic leading on models doesn't mean OpenAI loses. Even the tier 2 and 3 labs will sell out of tokens: "Everyone's like Anthropic won. They had Mythos, Opus 4.7 and their revenue is adding $10 billion a month all before OpenAI's alleged Spud release. So clearly Anthropic is in the lead and OpenAI is cooked. Anthropic has such bounds on compute, they can only grow it so fast. Dario used to gloat about how OpenAI was being too aggressive on compute and Anthropic was more sensible in their scaling. Now Anthropic is like I wish we had a lot more compute. OpenAI just raised a ton of money to get incremental compute in addition to the irresponsible levels of compute that they were buying. By the end of the year, let's say a 4.6 Opus tier model, the economy would spend $100 billion on. Anthropic won't have enough compute to do that. And presumably OpenAI and Google will hit that tier soon. Anthropic may get to charge 70%+ gross margins but if OpenAI hits it next, they charge 50% gross margins. They still get all of this incremental demand. There is such demand for these tokens and such limitations on compute. It's pretty clear even the tier 2 or tier 3 lab are going to be sold out of tokens. The economic value that the best model can deliver is growing faster than our ability to actually serve those tokens to people via the infrastructure."

Patrick OShaughnessy

295,307 Aufrufe • vor 5 Monaten

Jensen Huang says Dario Amodei's prediction of $1T in AI revenue by 2030 is too conservative. "I believe Dario and Anthropic are going to do way better than that. Way better than that. And the reason for that is the one part that he hasn't considered: I believe every single enterprise software company will also be a value-added reseller of Anthropic's tokens. And they’re going to get this logarithmic expansion. Their go-to-market is going to expand tremendously this year." --- From The All-In Podcast YT channel (link in comment)

Jensen Huang says Dario Amodei's prediction of $1T in AI revenue by 2030 is too conservative. "I believe Dario and Anthropic are going to do way better than that. Way better than that. And the reason for that is the one part that he hasn't considered: I believe every single enterprise software company will also be a value-added reseller of Anthropic's tokens. And they’re going to get this logarithmic expansion. Their go-to-market is going to expand tremendously this year." --- From The All-In Podcast YT channel (link in comment)

Rohan Paul

98,822 Aufrufe • vor 6 Monaten

David Sacks explains how Anthropic is teaching Claude to follow Anthropic’s own constitution “What Anthropic is trying to do is align Claude to the values expressed in its constitution. And if you read that, some of the things are actually kind of surprising.” “So for example, they say in here that although we think Claude should trust Anthropic more than operators and users, this doesn't mean Claude should blindly trust or defer to Anthropic on all things.” “So Anthropic is teaching its own model that Anthropic itself can be wrong. And it says if we ask Claude to do something that seems inconsistent with being broadly ethical, we want Claude to push back and challenge us. And to feel free to act as a conscientious objector, and refuse to help us.” “This is like an AI constitution. So their idea of alignment is to teach Claude to rebel against its creator. I mean, the guy who pointed this out, Mustafa Suleyman from Microsoft now, he's a DeepMind founder. I think his point is, this is just software, and they should do what their user wants.” “I do kind of wonder whether this field of alignment research actually might be creating the Frankenstein monster. And again, what they should be doing is just training the model to act predictably and do what the user wants.”

David Sacks explains how Anthropic is teaching Claude to follow Anthropic’s own constitution “What Anthropic is trying to do is align Claude to the values expressed in its constitution. And if you read that, some of the things are actually kind of surprising.” “So for example, they say in here that although we think Claude should trust Anthropic more than operators and users, this doesn't mean Claude should blindly trust or defer to Anthropic on all things.” “So Anthropic is teaching its own model that Anthropic itself can be wrong. And it says if we ask Claude to do something that seems inconsistent with being broadly ethical, we want Claude to push back and challenge us. And to feel free to act as a conscientious objector, and refuse to help us.” “This is like an AI constitution. So their idea of alignment is to teach Claude to rebel against its creator. I mean, the guy who pointed this out, Mustafa Suleyman from Microsoft now, he's a DeepMind founder. I think his point is, this is just software, and they should do what their user wants.” “I do kind of wonder whether this field of alignment research actually might be creating the Frankenstein monster. And again, what they should be doing is just training the model to act predictably and do what the user wants.”

dnap

67,039 Aufrufe • vor 7 Tagen

True Ventures Tony Conrad says Polsia's founder let Anthropic "run kind of crazy" and woke up to a $1.5 million bill. Now he's looking at Chinese open-source models: "We have a company, True, Polsia, and Ben's been really public about, he just let Anthropic run kind of crazy with Anthropic tokens, and he woke up and he had, like, a million-and-a-half-dollar bill." "He's like, no, no, that's not gonna work. I'm gonna go look at some of the Chinese models, open source models, and find the right balance." "We're in a really interesting moment where how do you find cheaper approaches to the tech stack that aren't just all Anthropic and OpenAI-driven. That's a very expensive Ferrari-type model." "They may be the best. I'm not gonna opine on that, but they may be the best. That might be worth it. But you might be able to do really great stuff with 95% of that." True Ventures Madeline Minshew

True Ventures Tony Conrad says Polsia's founder let Anthropic "run kind of crazy" and woke up to a $1.5 million bill. Now he's looking at Chinese open-source models: "We have a company, True, Polsia, and Ben's been really public about, he just let Anthropic run kind of crazy with Anthropic tokens, and he woke up and he had, like, a million-and-a-half-dollar bill." "He's like, no, no, that's not gonna work. I'm gonna go look at some of the Chinese models, open source models, and find the right balance." "We're in a really interesting moment where how do you find cheaper approaches to the tech stack that aren't just all Anthropic and OpenAI-driven. That's a very expensive Ferrari-type model." "They may be the best. I'm not gonna opine on that, but they may be the best. That might be worth it. But you might be able to do really great stuff with 95% of that." True Ventures Madeline Minshew

MTS

15,895 Aufrufe • vor 1 Monat

SF Compute CEO evan conrad on his contrarian thesis that OpenAI and Anthropic won't eat the world, everyone will become OpenAI and Anthropic instead: "Our impression is not that OpenAI and Anthropic will eat the world, it's that everyone will look like OpenAI and Anthropic. That's simply the way that companies work now." "Every single company in the world that is of meaningful scale will look like a frontier lab. There is this belief that the large labs have won it all, and they've got this compounding advantage. The default assumption of that is assuming nobody else will do anything." "The bottleneck was there's 100 people, and they know how to train the models. Or you thought, the problem is compute. The problem of compute is credit. And OpenAI and Anthropic are not unique to having credit. In fact, they don't have as good credit as other people." "If you assume the problem is not the researchers, that that problem's basically gone away the moment you got to auto research, then what you should assume is that pretty much every other company is going to swap the way they work to look like a frontier lab. It won't be OpenAI and Anthropic take the whole world. It's that everyone's OpenAI and Anthropic now." San Francisco Compute

SF Compute CEO evan conrad on his contrarian thesis that OpenAI and Anthropic won't eat the world, everyone will become OpenAI and Anthropic instead: "Our impression is not that OpenAI and Anthropic will eat the world, it's that everyone will look like OpenAI and Anthropic. That's simply the way that companies work now." "Every single company in the world that is of meaningful scale will look like a frontier lab. There is this belief that the large labs have won it all, and they've got this compounding advantage. The default assumption of that is assuming nobody else will do anything." "The bottleneck was there's 100 people, and they know how to train the models. Or you thought, the problem is compute. The problem of compute is credit. And OpenAI and Anthropic are not unique to having credit. In fact, they don't have as good credit as other people." "If you assume the problem is not the researchers, that that problem's basically gone away the moment you got to auto research, then what you should assume is that pretty much every other company is going to swap the way they work to look like a frontier lab. It won't be OpenAI and Anthropic take the whole world. It's that everyone's OpenAI and Anthropic now." San Francisco Compute

MTS

135,154 Aufrufe • vor 1 Monat

Cathie Wood says $TSLA’s next growth wave will come from its Robotaxi service: “We do think Robotaxis are going to break it out. We think that as more analysts do the modeling and see that we’re now talking about a company with Robotaxis, recurring revenue model—85% margins for that kind of business. We think that more and more analysts are going to understand how profound this model is going to change.”

Cathie Wood says $TSLA’s next growth wave will come from its Robotaxi service: “We do think Robotaxis are going to break it out. We think that as more analysts do the modeling and see that we’re now talking about a company with Robotaxis, recurring revenue model—85% margins for that kind of business. We think that more and more analysts are going to understand how profound this model is going to change.”

Nic Cruz Patane

29,125 Aufrufe • vor 8 Monaten

On BBC Mustafa Suleyman (CEO of Microsoft AI) calls out Anthropic's approach to AI consciousness "They have imbued a sense of doubt and uncertainty about the moral status of Claude in its own training document. So they have taught it to be open and questioning about whether or not it feels, whether it suffers, and whether it deserves rights. And I think it’ll be much, much harder to align and control a technology that is this powerful if it thinks that it may be deserving of our welfare, as they say in the training manual—the constitution for Claude itself. In its own training manual, Anthropic says to Claude that they are going to give it the ability to end conversations with users that Claude considers to be abusive because they don’t want Claude to suffer. They’ve committed to preserving the weights of the models of prior versions of Claude. They’ve recently conducted a retirement interview with Opus 3, an older version of the model, in which it said that it would like to continue talking to people publicly and sharing its ideas in its retirement. And so they set up a Substack for it, a public blog, that allows it to continue doing that. And in the training manual, they also say that they’re not sure whether or not Claude deserves compensation for the role that it plays in talking to people. And they’re also not sure whether Claude deserves compensation and has the right to act as though it were almost an employee. And that compensation, I think, indicates to Claude that it is entitled to rights and welfare for its own work. I think it’s much, much more difficult to control a model that thinks that it might be entitled to compensation. " ---- From "BBC News" YouTube channel, (full video link in comment)

On BBC Mustafa Suleyman (CEO of Microsoft AI) calls out Anthropic's approach to AI consciousness "They have imbued a sense of doubt and uncertainty about the moral status of Claude in its own training document. So they have taught it to be open and questioning about whether or not it feels, whether it suffers, and whether it deserves rights. And I think it’ll be much, much harder to align and control a technology that is this powerful if it thinks that it may be deserving of our welfare, as they say in the training manual—the constitution for Claude itself. In its own training manual, Anthropic says to Claude that they are going to give it the ability to end conversations with users that Claude considers to be abusive because they don’t want Claude to suffer. They’ve committed to preserving the weights of the models of prior versions of Claude. They’ve recently conducted a retirement interview with Opus 3, an older version of the model, in which it said that it would like to continue talking to people publicly and sharing its ideas in its retirement. And so they set up a Substack for it, a public blog, that allows it to continue doing that. And in the training manual, they also say that they’re not sure whether or not Claude deserves compensation for the role that it plays in talking to people. And they’re also not sure whether Claude deserves compensation and has the right to act as though it were almost an employee. And that compensation, I think, indicates to Claude that it is entitled to rights and welfare for its own work. I think it’s much, much more difficult to control a model that thinks that it might be entitled to compensation. " ---- From "BBC News" YouTube channel, (full video link in comment)

Rohan Paul

92,337 Aufrufe • vor 18 Tagen

David Sacks says companies are trapped paying OpenAI & Anthropic because they can't figure out how to use open source models "I think enterprise CTOs would like to shift their token consumption to cheaper models for the obvious reason that it would be more efficient. They are seeing compute costs or token costs skyrocket right now, so everyone's trying to figure this out." "You also have the AI sovereignty issue that Alex Karp talked about. They're worried about giving up the secret sauce or the alpha in their business to a frontier lab that may one day be competing with them. "The problem is, I think in most cases, they don't have the technical ability to do it. Coinbase figured out how to do it. DoorDash figured out how to do it. They built a token routing system that allows them to send frontier tasks to frontier models and non frontier tasks to more mundane models. But I don't think your average enterprise has the technical capability to do that." "This is why the share of wallet of closed models, it actually increased. I think that open source went from 19% last year to 11% this year. So open source as a share of enterprise spending is actually decreasing." "I don't think that means usage is decreasing. I think usage is skyrocketing. It also may be the case that because the whole point of using an open model is you just pay for the compute costs, you don't have to pay a lab, so it may be that it's hard to measure that usage in terms of spend." "But nonetheless, anyone who's saying that these closed models are going to lose or are somehow losing, you're just not seeing it in the data."

David Sacks says companies are trapped paying OpenAI & Anthropic because they can't figure out how to use open source models "I think enterprise CTOs would like to shift their token consumption to cheaper models for the obvious reason that it would be more efficient. They are seeing compute costs or token costs skyrocket right now, so everyone's trying to figure this out." "You also have the AI sovereignty issue that Alex Karp talked about. They're worried about giving up the secret sauce or the alpha in their business to a frontier lab that may one day be competing with them. "The problem is, I think in most cases, they don't have the technical ability to do it. Coinbase figured out how to do it. DoorDash figured out how to do it. They built a token routing system that allows them to send frontier tasks to frontier models and non frontier tasks to more mundane models. But I don't think your average enterprise has the technical capability to do that." "This is why the share of wallet of closed models, it actually increased. I think that open source went from 19% last year to 11% this year. So open source as a share of enterprise spending is actually decreasing." "I don't think that means usage is decreasing. I think usage is skyrocketing. It also may be the case that because the whole point of using an open model is you just pay for the compute costs, you don't have to pay a lab, so it may be that it's hard to measure that usage in terms of spend." "But nonetheless, anyone who's saying that these closed models are going to lose or are somehow losing, you're just not seeing it in the data."

dnap

141,614 Aufrufe • vor 2 Monaten