Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Sundar Pichai just dropped a surprisingly candid look into Google DeepMind’s current AI heartbeat. Here’s the essence of the clip: • Google is now operating on a steady six month cadence, pushing the frontier twice a year • Sundar openly admits that meaningful leaps are getting harder • Gemini...

VraserX's profile picture

VraserX e/acc

20,882 subscribers

162,001 görüntüleme • 9 ay önce •via X (Twitter)

EğitimHaberler & PolitikaBilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Google isn’t betting on a single AI architecture. Sundar Pichai, CEO of Google: “We’re going to push the diffusion paradigm as hard as possible.” “All of today’s mainline Gemini models are autoregressive. Diffusion is a different paradigm.” “For the same capability, diffusion can be much faster.” “It’s behind the mainline models today, but there will be areas where it’s the right tool.” “We’re pushing multiple directions in parallel, and bringing them together where it makes sense.”
1:02

Google isn’t betting on a single AI architecture. Sundar Pichai, CEO of Google: “We’re going to push the diffusion paradigm as hard as possible.” “All of today’s mainline Gemini models are autoregressive. Diffusion is a different paradigm.” “For the same capability, diffusion can be much faster.” “It’s behind the mainline models today, but there will be areas where it’s the right tool.” “We’re pushing multiple directions in parallel, and bringing them together where it makes sense.”

Forward Future

152,933 görüntüleme • 8 ay önce

Sundar Pichai on the robotics opportunity: ⦿ He's impressed by recent humanoid progress - so much so that he sometimes has to look closely to tell if a robot video is real or fake. ⦿ Google initially moved too early into the application layer, but now sees the combination of AI and robotics as the “sweet spot.” ⦿ He believes we’re 2–3 years away from a “magical moment” in robotics. ⦿ Google is working on playing a role similar to Android in robotics - one of its internal bets is already focused on supporting robotics manufacturers. ⦿ Gemini robotics models will work very well - how to productize them is still an open question at the company.
2:43

Sundar Pichai on the robotics opportunity: ⦿ He's impressed by recent humanoid progress - so much so that he sometimes has to look closely to tell if a robot video is real or fake. ⦿ Google initially moved too early into the application layer, but now sees the combination of AI and robotics as the “sweet spot.” ⦿ He believes we’re 2–3 years away from a “magical moment” in robotics. ⦿ Google is working on playing a role similar to Android in robotics - one of its internal bets is already focused on supporting robotics manufacturers. ⦿ Gemini robotics models will work very well - how to productize them is still an open question at the company.

The Humanoid Hub

87,088 görüntüleme • 1 yıl önce

SUNDAR PICHAI: AI IS REWRITING EARTH’S INFRASTRUCTURE IN TWO YEARS Sundar laid it out with quiet certainty: the AI boom isn’t coming, it’s already rewriting the laws of scale: "You know, maybe four years ago Google was spending less than $30 billion per year. This year, that number is gonna be over $90 billion. And if you collectively add what all the companies are doing, we have well over a trillion dollars of investment going in, in building the infrastructure for this moment. And one way I think about it is in the next couple of years, we'll end up building what we probably built in the past 10 to 20 years. So hopefully just in a couple of years. So that gives you the scale at which this is ramping up." It’s the kind of scale that makes empires blink. Alphabet’s AI capex shot from $24 billion in 2021 to over $90 billion in 2025, powered by Gemini, Vertex AI, and an arms race for compute that’s now burning through a trillion dollars across Google, Microsoft, Amazon, and Meta. Source: thewizeai / Sundar Pichai
0:36

SUNDAR PICHAI: AI IS REWRITING EARTH’S INFRASTRUCTURE IN TWO YEARS Sundar laid it out with quiet certainty: the AI boom isn’t coming, it’s already rewriting the laws of scale: "You know, maybe four years ago Google was spending less than $30 billion per year. This year, that number is gonna be over $90 billion. And if you collectively add what all the companies are doing, we have well over a trillion dollars of investment going in, in building the infrastructure for this moment. And one way I think about it is in the next couple of years, we'll end up building what we probably built in the past 10 to 20 years. So hopefully just in a couple of years. So that gives you the scale at which this is ramping up." It’s the kind of scale that makes empires blink. Alphabet’s AI capex shot from $24 billion in 2021 to over $90 billion in 2025, powered by Gemini, Vertex AI, and an arms race for compute that’s now burning through a trillion dollars across Google, Microsoft, Amazon, and Meta. Source: thewizeai / Sundar Pichai

Mario Nawfal

163,412 görüntüleme • 9 ay önce

How did a tiny, scrappy team build one of the most powerful AI voice models? In a deep dive with Sesame CTO Ankit Kumar and a16z's Anjney Midha, we explore how Sesame is pushing the boundaries of AI conversation, why it open-sourced its speech generation model, and the power of small teams to outdo much larger AI labs on product focus. A part of their secret: a relentless focus on real-time, natural conversations over raw intelligence, and a deep commitment to voice, personality, and user experience. By opening up its speech generation model, Sesame is paving the way for even more breakthroughs in AI-native conversation 👇
1:04

How did a tiny, scrappy team build one of the most powerful AI voice models? In a deep dive with Sesame CTO Ankit Kumar and a16z's Anjney Midha, we explore how Sesame is pushing the boundaries of AI conversation, why it open-sourced its speech generation model, and the power of small teams to outdo much larger AI labs on product focus. A part of their secret: a relentless focus on real-time, natural conversations over raw intelligence, and a deep commitment to voice, personality, and user experience. By opening up its speech generation model, Sesame is paving the way for even more breakthroughs in AI-native conversation 👇

a16z

29,930 görüntüleme • 1 yıl önce

Chinese AI models are wiping billions off Big Tech right now. Google just lost $200 billion in a single day, and the model it needed to fight back still isn't ready. Gemini 3.5 Pro, Google's most powerful model, is months behind schedule. Alphabet stock dropped 4.4% that same day. The Deepseek moment is happening again, and the new model is FAR bigger. On the same day Google's delay leaked, a Beijing lab called Moonshot released Kimi K3. It is the largest open model ever built, with 2.8 trillion parameters. It took the number one spot on the Frontend Code Arena, a live coding leaderboard, passing Anthropic's best model. And Moonshot is giving it away for free on July 27. The genius part: Anyone with enough computers can download it and run a frontier level AI without paying a cent to a US company. A single task on Kimi K3 costs about 94 cents. The same work on some American models costs nearly double. So why would a company keep paying premium prices for a model it can now get for free? The entire US AI business is built on selling access to models that cost billions to train. If a free Chinese version does most of the same work, that pricing power starts to crack. And Kimi is close to the best. On one closely watched intelligence ranking it scored 57, just behind the top American models GPT-5.6 Sol and Fable 5, and ahead of Claude Opus 4.8. Bank of America told clients that Kimi proves Chinese labs can keep making big leaps even with limited chips. And the founder of Moonshot, Yang Zhilin, learned to build AI as a researcher INSIDE Google. Google literally wrote the 2017 paper that made all of these models possible. Now the people who studied its work are using it to destroy Google, and handing it out for free. What happens next: Kimi K3's weights go public on July 27. Google reports earnings on July 22, and everyone will be asking the same question about Gemini. If free models keep topping the charts, every valuation built on paid AI access has to be rewritten. What do you think?
1:10

Chinese AI models are wiping billions off Big Tech right now. Google just lost $200 billion in a single day, and the model it needed to fight back still isn't ready. Gemini 3.5 Pro, Google's most powerful model, is months behind schedule. Alphabet stock dropped 4.4% that same day. The Deepseek moment is happening again, and the new model is FAR bigger. On the same day Google's delay leaked, a Beijing lab called Moonshot released Kimi K3. It is the largest open model ever built, with 2.8 trillion parameters. It took the number one spot on the Frontend Code Arena, a live coding leaderboard, passing Anthropic's best model. And Moonshot is giving it away for free on July 27. The genius part: Anyone with enough computers can download it and run a frontier level AI without paying a cent to a US company. A single task on Kimi K3 costs about 94 cents. The same work on some American models costs nearly double. So why would a company keep paying premium prices for a model it can now get for free? The entire US AI business is built on selling access to models that cost billions to train. If a free Chinese version does most of the same work, that pricing power starts to crack. And Kimi is close to the best. On one closely watched intelligence ranking it scored 57, just behind the top American models GPT-5.6 Sol and Fable 5, and ahead of Claude Opus 4.8. Bank of America told clients that Kimi proves Chinese labs can keep making big leaps even with limited chips. And the founder of Moonshot, Yang Zhilin, learned to build AI as a researcher INSIDE Google. Google literally wrote the 2017 paper that made all of these models possible. Now the people who studied its work are using it to destroy Google, and handing it out for free. What happens next: Kimi K3's weights go public on July 27. Google reports earnings on July 22, and everyone will be asking the same question about Gemini. If free models keep topping the charts, every valuation built on paid AI access has to be rewritten. What do you think?

Ricardo

47,790 görüntüleme • 1 ay önce

Former Google CEO Eric Schmidt says a frontier AI model costs up to $500 million to build and is worth about six months: "Are the scaling laws stopping?" "And the answer right now is no, which is unbelievable." "A model costs $200 to $500 million, you release it, and it has a life expectancy of about six months." "And then there's another one." "And the competition between the leading players is so brutal, right, that they literally count in the number of weeks of value of these things, which costs hundreds of millions of dollars." Schmidt is talking about the build cost. There is a second cost underneath it, which is re-proving each replacement, because whatever testing cleared one version does not carry over to the next. A company that standardizes on a frontier model this year is signing up to do that re-approval work twice a year. - Eric Schmidt (Eric Schmidt), former CEO of Google (Google) and now Chair and CEO of Relativity Space, with Alan Blinder at Princeton's Griswold Center (Princeton University).
0:34

Former Google CEO Eric Schmidt says a frontier AI model costs up to $500 million to build and is worth about six months: "Are the scaling laws stopping?" "And the answer right now is no, which is unbelievable." "A model costs $200 to $500 million, you release it, and it has a life expectancy of about six months." "And then there's another one." "And the competition between the leading players is so brutal, right, that they literally count in the number of weeks of value of these things, which costs hundreds of millions of dollars." Schmidt is talking about the build cost. There is a second cost underneath it, which is re-proving each replacement, because whatever testing cleared one version does not carry over to the next. A company that standardizes on a frontier model this year is signing up to do that re-approval work twice a year. - Eric Schmidt (Eric Schmidt), former CEO of Google (Google) and now Chair and CEO of Relativity Space, with Alan Blinder at Princeton's Griswold Center (Princeton University).

Karl Mehta

492,962 görüntüleme • 26 gün önce

Sundar Pichai just reminded everyone that Google was built on open source. He personally worked on Chromium, Android, and Kubernetes before becoming CEO. Three systems that power billions of devices today. Now Google is applying the same philosophy to AI. Their Gemma models are updated year after year, designed to run on edge devices rather than requiring massive cloud infrastructure. The question he got was why not release a large open source frontier model. His answer was revealing: frontier training requires enormous capex investment, and Google is putting billions into R&D to stay at the frontier. The open source AI landscape is quietly consolidating around a few players who can sustain the investment. Google's two decades of open source credibility gives them a position that's hard to replicate.
2:19

Sundar Pichai just reminded everyone that Google was built on open source. He personally worked on Chromium, Android, and Kubernetes before becoming CEO. Three systems that power billions of devices today. Now Google is applying the same philosophy to AI. Their Gemma models are updated year after year, designed to run on edge devices rather than requiring massive cloud infrastructure. The question he got was why not release a large open source frontier model. His answer was revealing: frontier training requires enormous capex investment, and Google is putting billions into R&D to stay at the frontier. The open source AI landscape is quietly consolidating around a few players who can sustain the investment. Google's two decades of open source credibility gives them a position that's hard to replicate.

Shruti

152,324 görüntüleme • 1 ay önce

Google just pulled off the biggest theft in the history of AI. And their OWN documents exposed it... Three of the largest publishers on Earth, Hachette, Cengage, and Elsevier, just sued Google in federal court, alongside best-selling author Scott Turow. Their claim is that Google built Gemini, its flagship AI, on millions of copyrighted works it never paid for or licensed. This is one of the biggest copyright cases ever aimed at an AI company: Before they trained Gemini, an internal Google document allegedly spelled out the risk directly. Using this material could expose the company to "$10Bs-$100Bs in potential fines." Google's own people put a number on the theft, in the tens of billions, but the company moved ahead anyway. Then it allegedly tried to DELETE the evidence... The complaint says Google stripped the copyright information off the works before feeding them into Gemini, so nobody could trace what the model had actually been trained on. Pull the fingerprints off first, and the theft gets much harder to prove later. And there's a second betrayal underneath the first: Most of this material was not scraped from some random corner of the internet. Publishers had handed it to Google years earlier for a narrow purpose, to make their catalogs searchable inside Google's own services. The lawsuit says Google took that trust and repurposed the content to build a machine that now competes directly with the people who supplied it. And that machine is the entire point. The complaint describes Gemini producing a full-length substitute for a copyrighted work in about 20 minutes. Something an author spent years writing can now be cloned in an afternoon by the company that trained on the original. No writer or publisher survives that. So why does this case matter more than the dozen other AI copyright fights? Because most of them turn on a fuzzy fair-use question argued years after the fact. This one arrives with an internal document that allegedly SHOWS Google weighed the cost of getting caught and trained on the material anyway. A jury does not need a law degree to read that. The fight now moves toward discovery, where Google's internal emails and training records get dragged into the open. If those files back up what the complaint claims, that $100 billion will turn into a real liability. Google spent years telling the world it was organizing information for everyone. Its own documents show it knew exactly whose information it was taking, and what the price of getting caught would be.
1:15

Google just pulled off the biggest theft in the history of AI. And their OWN documents exposed it... Three of the largest publishers on Earth, Hachette, Cengage, and Elsevier, just sued Google in federal court, alongside best-selling author Scott Turow. Their claim is that Google built Gemini, its flagship AI, on millions of copyrighted works it never paid for or licensed. This is one of the biggest copyright cases ever aimed at an AI company: Before they trained Gemini, an internal Google document allegedly spelled out the risk directly. Using this material could expose the company to "$10Bs-$100Bs in potential fines." Google's own people put a number on the theft, in the tens of billions, but the company moved ahead anyway. Then it allegedly tried to DELETE the evidence... The complaint says Google stripped the copyright information off the works before feeding them into Gemini, so nobody could trace what the model had actually been trained on. Pull the fingerprints off first, and the theft gets much harder to prove later. And there's a second betrayal underneath the first: Most of this material was not scraped from some random corner of the internet. Publishers had handed it to Google years earlier for a narrow purpose, to make their catalogs searchable inside Google's own services. The lawsuit says Google took that trust and repurposed the content to build a machine that now competes directly with the people who supplied it. And that machine is the entire point. The complaint describes Gemini producing a full-length substitute for a copyrighted work in about 20 minutes. Something an author spent years writing can now be cloned in an afternoon by the company that trained on the original. No writer or publisher survives that. So why does this case matter more than the dozen other AI copyright fights? Because most of them turn on a fuzzy fair-use question argued years after the fact. This one arrives with an internal document that allegedly SHOWS Google weighed the cost of getting caught and trained on the material anyway. A jury does not need a law degree to read that. The fight now moves toward discovery, where Google's internal emails and training records get dragged into the open. If those files back up what the complaint claims, that $100 billion will turn into a real liability. Google spent years telling the world it was organizing information for everyone. Its own documents show it knew exactly whose information it was taking, and what the price of getting caught would be.

Ricardo

26,108 görüntüleme • 1 ay önce

Sundar Pichai just confirmed that nobody on earth is using Google’s best model. Including him. Pichai: “The models we all use the most is maybe like a few months behind the maximum capability we can deliver.” Finished intelligence is sitting in reserve, built and working, waiting for the economics to catch up. What reaches you is not the edge of research. It is the edge of what can be served to billions of people without losing money on every request. Pichai: “for each generation, we feel like we’ve been able to get the Pro model at like, I don’t know, 80-90% of Ultra’s capability.” Google stopped shipping Ultra. The capability existed. Serving it did not. Pichai: “But what we’ve been able to do is to go to the next generation and make the next generation’s Pro as good as the previous generation’s Ultra.” The frontier gets built, held back, compressed, and released a generation later at a price that works. Every efficiency gain drains part of that backlog at once, which is why progress keeps arriving in jumps that feel larger than the research behind them. You are not watching discovery. You are watching a queue clear. Benchmarks stopped tracking any of this, because they measure the ceiling and almost nobody works at the ceiling. Fridman: “benchmarks are less and less capable of capturing the intelligence of models, the effectiveness of models.” A model that is slightly less capable and dramatically faster wins nearly every real task, because latency decides what you are willing to ask in the first place. Fridman: “you could argue Gemini Flash is much more impactful than Pro. Just because of the latency, it’s super intelligent already.” Wait thirty seconds for an answer and you ask a few questions a day. Get it back instantly and you rebuild your workflow around it. Value is created at the point of use, never at the top of a leaderboard. Intelligence stopped being the scarce input. Delivery became the constraint, and delivery is an engineering problem, which is the one category of problem humans have never lost. Electricity was real for decades before it reached a kitchen. The invention finishes early. Distribution takes longer, and it always gets built. Scaling did not stall. What closed is the gap between what exists and what reaches you. The best model in the world has already been built. What is left is a cost curve, and cost curves only run one direction.
1:47

Sundar Pichai just confirmed that nobody on earth is using Google’s best model. Including him. Pichai: “The models we all use the most is maybe like a few months behind the maximum capability we can deliver.” Finished intelligence is sitting in reserve, built and working, waiting for the economics to catch up. What reaches you is not the edge of research. It is the edge of what can be served to billions of people without losing money on every request. Pichai: “for each generation, we feel like we’ve been able to get the Pro model at like, I don’t know, 80-90% of Ultra’s capability.” Google stopped shipping Ultra. The capability existed. Serving it did not. Pichai: “But what we’ve been able to do is to go to the next generation and make the next generation’s Pro as good as the previous generation’s Ultra.” The frontier gets built, held back, compressed, and released a generation later at a price that works. Every efficiency gain drains part of that backlog at once, which is why progress keeps arriving in jumps that feel larger than the research behind them. You are not watching discovery. You are watching a queue clear. Benchmarks stopped tracking any of this, because they measure the ceiling and almost nobody works at the ceiling. Fridman: “benchmarks are less and less capable of capturing the intelligence of models, the effectiveness of models.” A model that is slightly less capable and dramatically faster wins nearly every real task, because latency decides what you are willing to ask in the first place. Fridman: “you could argue Gemini Flash is much more impactful than Pro. Just because of the latency, it’s super intelligent already.” Wait thirty seconds for an answer and you ask a few questions a day. Get it back instantly and you rebuild your workflow around it. Value is created at the point of use, never at the top of a leaderboard. Intelligence stopped being the scarce input. Delivery became the constraint, and delivery is an engineering problem, which is the one category of problem humans have never lost. Electricity was real for decades before it reached a kitchen. The invention finishes early. Distribution takes longer, and it always gets built. Scaling did not stall. What closed is the gap between what exists and what reaches you. The best model in the world has already been built. What is left is a cost curve, and cost curves only run one direction.

Dustin

125,137 görüntüleme • 26 gün önce

DeepMind CEO Demis Hassabis says the AI bottleneck is already moving from model ideas to physical supply chains. "How much of the restriction at this point is physical? I am talking memory shortage, GPU shortage, and electricity." Demis: "That is constraining a lot of the deployment for sure. We see so much more demand for our models - Gemini and the other models that we have - than we can possibly serve." Then the less obvious part: "It does constrain a little bit the research, because you need a lot of chips to be able to experiment on new ideas at big enough scale that you can actually see if they are going to work." And on memory: "The whole supply chain is kind of strained... It still in the end actually comes down to a few suppliers of a few of the key components. Anywhere where there is some kind of constraint on the capacity, there is a sort of choke point." Important framing: The next AI race is not just who has the best model architecture. It is who can secure memory, chips, power, suppliers, and enough experimental capacity to find the next breakthrough before everyone else.
1:09

DeepMind CEO Demis Hassabis says the AI bottleneck is already moving from model ideas to physical supply chains. "How much of the restriction at this point is physical? I am talking memory shortage, GPU shortage, and electricity." Demis: "That is constraining a lot of the deployment for sure. We see so much more demand for our models - Gemini and the other models that we have - than we can possibly serve." Then the less obvious part: "It does constrain a little bit the research, because you need a lot of chips to be able to experiment on new ideas at big enough scale that you can actually see if they are going to work." And on memory: "The whole supply chain is kind of strained... It still in the end actually comes down to a few suppliers of a few of the key components. Anywhere where there is some kind of constraint on the capacity, there is a sort of choke point." Important framing: The next AI race is not just who has the best model architecture. It is who can secure memory, chips, power, suppliers, and enough experimental capacity to find the next breakthrough before everyone else.

Karl Mehta

25,716 görüntüleme • 2 ay önce

Google just quit the AI race on purpose, and it is about to make MORE money than everyone still running it. 4 of the most cited AI researchers alive walked out of Google in a single afternoon. Jeff Dean, the man who built the systems Google runs on, gone after 27 years. Sanjay Ghemawat, his longtime partner, gone. Oriol Vinyals, a Gemini co-lead, gone. Quoc Le, a Google Brain co-founder, gone. That same day, Demis Hassabis stepped back from running DeepMind. Hassabis co-founded the lab, won a Nobel Prize for AlphaFold, and had been the face of Google AI for a decade. The stock dropped 5% within hours. Analysts called it a brain drain. Headlines called it the day Google fell behind. But turns out that's completely wrong, because the numbers underneath tell a completely different story: Google is not trying to win the frontier model race anymore. It looked at where the money is and walked toward it. Gemini, Google's flagship model business, generated about $12 billion in annual revenue last quarter. That is the entire payoff from competing head to head with OpenAI and Anthropic. Now look at the other number. By the end of 2027, Google Cloud is projected to do over $73 billion selling AI infrastructure to other companies, plus another $120 billion selling its TPU chips. That is roughly $200 billion of external sales at high margins, against a $12 billion model business. Google understood that the frontier race is the expensive part while selling the shovels is the profitable part. And the customers buying those shovels include Google's own rivals. Over 20% of Google's TPU shipments for 2026 and 2027 are going to Anthropic, one of the two labs supposedly beating Gemini. Google now makes money every time Anthropic trains a model designed to crush Google's OWN product. Cede the frontier, own the layer underneath it, and collect a toll from everyone racing across the top. The researchers leaving is the symptom of a company that already decided models are not where it wins. Jeff Dean said it himself on the way out. He told the New York Times that leaving a public company gives him room to make decisions "not necessarily in the company's purist financial interests." Read that from Google's side: The people who wanted to chase the science left, because Google is now optimizing for the FINANCIAL interest. Gemini 3.5 Pro is running months behind, with staff blaming low morale. DeepMind's comms, legal, and marketing teams are being folded into Google proper. A former manager told the Guardian the era of DeepMind as an independent lab is over. None of that reads as failure once you see the strategy. Yet Wall Street is pricing this as Google losing. The parallel that should worry the frontier labs: If open weight models keep compressing the price of inference, being the best model stops being a business. It becomes like semiconductor fabrication, strategically vital and financially brutal, a race you win and still lose money running. Google is the first giant to admit that. The company that invented the transformer just handed the frontier to OpenAI and Anthropic, and positioned itself to get paid on every model both of them ship. Those labs will be burning billions to stay one benchmark ahead, and Google will be cashing in hundreds of billions from it. The model business is actually just a race where everyone loses. Apple understood that from the get-go and never joined the race, Google understood it now and left it to OpenAI and Anthropic. Who will go bankrupt first?
2:05

Google just quit the AI race on purpose, and it is about to make MORE money than everyone still running it. 4 of the most cited AI researchers alive walked out of Google in a single afternoon. Jeff Dean, the man who built the systems Google runs on, gone after 27 years. Sanjay Ghemawat, his longtime partner, gone. Oriol Vinyals, a Gemini co-lead, gone. Quoc Le, a Google Brain co-founder, gone. That same day, Demis Hassabis stepped back from running DeepMind. Hassabis co-founded the lab, won a Nobel Prize for AlphaFold, and had been the face of Google AI for a decade. The stock dropped 5% within hours. Analysts called it a brain drain. Headlines called it the day Google fell behind. But turns out that's completely wrong, because the numbers underneath tell a completely different story: Google is not trying to win the frontier model race anymore. It looked at where the money is and walked toward it. Gemini, Google's flagship model business, generated about $12 billion in annual revenue last quarter. That is the entire payoff from competing head to head with OpenAI and Anthropic. Now look at the other number. By the end of 2027, Google Cloud is projected to do over $73 billion selling AI infrastructure to other companies, plus another $120 billion selling its TPU chips. That is roughly $200 billion of external sales at high margins, against a $12 billion model business. Google understood that the frontier race is the expensive part while selling the shovels is the profitable part. And the customers buying those shovels include Google's own rivals. Over 20% of Google's TPU shipments for 2026 and 2027 are going to Anthropic, one of the two labs supposedly beating Gemini. Google now makes money every time Anthropic trains a model designed to crush Google's OWN product. Cede the frontier, own the layer underneath it, and collect a toll from everyone racing across the top. The researchers leaving is the symptom of a company that already decided models are not where it wins. Jeff Dean said it himself on the way out. He told the New York Times that leaving a public company gives him room to make decisions "not necessarily in the company's purist financial interests." Read that from Google's side: The people who wanted to chase the science left, because Google is now optimizing for the FINANCIAL interest. Gemini 3.5 Pro is running months behind, with staff blaming low morale. DeepMind's comms, legal, and marketing teams are being folded into Google proper. A former manager told the Guardian the era of DeepMind as an independent lab is over. None of that reads as failure once you see the strategy. Yet Wall Street is pricing this as Google losing. The parallel that should worry the frontier labs: If open weight models keep compressing the price of inference, being the best model stops being a business. It becomes like semiconductor fabrication, strategically vital and financially brutal, a race you win and still lose money running. Google is the first giant to admit that. The company that invented the transformer just handed the frontier to OpenAI and Anthropic, and positioned itself to get paid on every model both of them ship. Those labs will be burning billions to stay one benchmark ahead, and Google will be cashing in hundreds of billions from it. The model business is actually just a race where everyone loses. Apple understood that from the get-go and never joined the race, Google understood it now and left it to OpenAI and Anthropic. Who will go bankrupt first?

Ricardo

241,927 görüntüleme • 24 gün önce

Sundar Pichai says Google's Antigravity built a working OS from scratch overnight. Then he named what comes next, the one thing he says no company should decide on alone: "You can see in Antigravity today... in over 12 hours, it can build a simple OS from scratch." "Genuinely those are multiple thousands of hours for somebody to do." "We are all in our products in some version or the other, have agents and subagents and the orchestration of those agents building things together. It's a continuum." "In the way people describe RSI, I don't think we are there yet... that would represent the next level of acceleration." "It shouldn't be an internal conversation at that point. I think it has to be a much broader conversation than that." "We all have to avoid race conditions at those stages of AGI." Twelve hours of machine time replaced what Pichai puts at thousands of hours of human engineering. The ramp is real. The step after it, models improving themselves, is one he thinks should not be decided inside a single company. Except right now it can only be decided inside a single company. No lab has to notify anyone before crossing that line. No outside test exists for it. The rest of us would find out afterward. Pichai is asking for a broader conversation that has no room to happen in yet. - Sundar Pichai (Sundar Pichai), CEO of Google and Alphabet, on Hard Fork with Casey Newton (Casey Newton) and Kevin Roose (Kevin Roose).
1:36

Sundar Pichai says Google's Antigravity built a working OS from scratch overnight. Then he named what comes next, the one thing he says no company should decide on alone: "You can see in Antigravity today... in over 12 hours, it can build a simple OS from scratch." "Genuinely those are multiple thousands of hours for somebody to do." "We are all in our products in some version or the other, have agents and subagents and the orchestration of those agents building things together. It's a continuum." "In the way people describe RSI, I don't think we are there yet... that would represent the next level of acceleration." "It shouldn't be an internal conversation at that point. I think it has to be a much broader conversation than that." "We all have to avoid race conditions at those stages of AGI." Twelve hours of machine time replaced what Pichai puts at thousands of hours of human engineering. The ramp is real. The step after it, models improving themselves, is one he thinks should not be decided inside a single company. Except right now it can only be decided inside a single company. No lab has to notify anyone before crossing that line. No outside test exists for it. The rest of us would find out afterward. Pichai is asking for a broader conversation that has no room to happen in yet. - Sundar Pichai (Sundar Pichai), CEO of Google and Alphabet, on Hard Fork with Casey Newton (Casey Newton) and Kevin Roose (Kevin Roose).

Karl Mehta

79,883 görüntüleme • 1 ay önce

Chamath: Tesla and Google are best positioned to win AI because they are building the full-stack Jason: "The game here is who wins the AI prize five years from now?" "Chamath, who's your number one and number two?" Chamath: " Tesla's one and Google's two." "And the reason is because they are the closest to having that vertically integrated stack that I spoke about." TESLA 1⃣ "I think that Tesla has the best vision models." "Now with xAI, they'll have one of the best LLMs and reasoning models." "And they'll be able to eventually stick that on Dojo." "And then all of that will be in all of the physical AI that you will interact with in your daily life." "Whether it's a robot, or whether it's a car, or whether it's a Robotaxi. So that's number one." GOOGLE 2⃣ "And then number two, for many of the same reasons, I think Google." "Because you'll have the Gemini family of models, which just absolutely kick ass." "Like Veo 3, which we haven't really spoke about, is gonna destroy Hollywood in the next year, like Hollywood is done, I think." "But they're landing model after model." "They have the TPU, and the next generation TPU I think is exceptional." "They're baking quantum." "And then they have an entire funnel of billions of people that they can direct experiences to." "So Tesla one, Google two."
1:04

Chamath: Tesla and Google are best positioned to win AI because they are building the full-stack Jason: "The game here is who wins the AI prize five years from now?" "Chamath, who's your number one and number two?" Chamath: " Tesla's one and Google's two." "And the reason is because they are the closest to having that vertically integrated stack that I spoke about." TESLA 1⃣ "I think that Tesla has the best vision models." "Now with xAI, they'll have one of the best LLMs and reasoning models." "And they'll be able to eventually stick that on Dojo." "And then all of that will be in all of the physical AI that you will interact with in your daily life." "Whether it's a robot, or whether it's a car, or whether it's a Robotaxi. So that's number one." GOOGLE 2⃣ "And then number two, for many of the same reasons, I think Google." "Because you'll have the Gemini family of models, which just absolutely kick ass." "Like Veo 3, which we haven't really spoke about, is gonna destroy Hollywood in the next year, like Hollywood is done, I think." "But they're landing model after model." "They have the TPU, and the next generation TPU I think is exceptional." "They're baking quantum." "And then they have an entire funnel of billions of people that they can direct experiences to." "So Tesla one, Google two."

The All-In Podcast

7,919,302 görüntüleme • 1 yıl önce

you told it to double check its work that is the line making it worse current models verify without being asked, so your instruction lands on top of behaviour that already happened, and what you get is the same task run twice, billed twice, reported at twice the length it reads like diligence it is a model doing a lap it had already finished every prompting guide ever written tells you to add that line on this generation you improve the output by deleting it
1:21

you told it to double check its work that is the line making it worse current models verify without being asked, so your instruction lands on top of behaviour that already happened, and what you get is the same task run twice, billed twice, reported at twice the length it reads like diligence it is a model doing a lap it had already finished every prompting guide ever written tells you to add that line on this generation you improve the output by deleting it

Lagosaur

14,673 görüntüleme • 15 gün önce

Pixel exists so Google can demonstrate its own services under conditions it controls, and in 2026 that service is Gemini. Anyone can run Gemini on a Galaxy or an iPhone. Only on Pixel does Google set the silicon, the integration, and the ship date together. The foldable comparisons for Pixel Fold might be unflattering today because of the excitement around Galaxy Fold 8 and the upcoming iPhone but Google has a full updated portfolio. Plus, Google can afford to be second on foldable geometry. It cannot afford to be second on how Gemini feels day to day.
1:14

Pixel exists so Google can demonstrate its own services under conditions it controls, and in 2026 that service is Gemini. Anyone can run Gemini on a Galaxy or an iPhone. Only on Pixel does Google set the silicon, the integration, and the ship date together. The foldable comparisons for Pixel Fold might be unflattering today because of the excitement around Galaxy Fold 8 and the upcoming iPhone but Google has a full updated portfolio. Plus, Google can afford to be second on foldable geometry. It cannot afford to be second on how Gemini feels day to day.

Carolina Milanesi

19,100 görüntüleme • 23 gün önce

AI video is getting wild… but THIS is on another level. 🤯🎬 My colleague Nick Lentz just dropped his submission for the Higgsfield Action Contest, and it’s honestly one of the most realistic and cinematic uses of AI video I’ve seen yet. The nuance of the characters. The action. The subtlety of the shots. It feels like something straight out of a film. If you’re into AI filmmaking or just want to see where this tech is heading, you’ve gotta check it out. If you like it, drop a comment or give it a thumbs up and show him some love. 🔥 Here's the link where you can vote:
2:53

AI video is getting wild… but THIS is on another level. 🤯🎬 My colleague Nick Lentz just dropped his submission for the Higgsfield Action Contest, and it’s honestly one of the most realistic and cinematic uses of AI video I’ve seen yet. The nuance of the characters. The action. The subtlety of the shots. It feels like something straight out of a film. If you’re into AI filmmaking or just want to see where this tech is heading, you’ve gotta check it out. If you like it, drop a comment or give it a thumbs up and show him some love. 🔥 Here's the link where you can vote:

Everett World

23,075 görüntüleme • 5 ay önce

Just write what you want to automate, and Todoist Automations takes care of the rest. A small preview of how we’re turning Todoist into something that does more of the heavy lifting for you (across all of your productivity apps). It may look simple on the surface, but there’s a lot of innovation and hard tech behind it. I’m very excited about where this is going!
1:59

Just write what you want to automate, and Todoist Automations takes care of the rest. A small preview of how we’re turning Todoist into something that does more of the heavy lifting for you (across all of your productivity apps). It may look simple on the surface, but there’s a lot of innovation and hard tech behind it. I’m very excited about where this is going!

Amir Salihefendić

20,269 görüntüleme • 5 ay önce