Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

SEO spent 20 years training its own replacement. Marketers did exactly what Google asked, answer the question better than anyone, and got so good at it that Google no longer needs to send you the visitor. It just reads your work out loud. This video breaks down why the... show more

Neil Patel

477,563 subscribers

15,347 görüntüleme • 2 ay önce •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

You can rank #1 on Google and still lose more than half your clicks. Most brands haven't noticed yet. For 25 years the internet ran on one deal: you give Google your content, Google sends you traffic. That deal is breaking. Our data shows organic click through rates falling by as much as 61%, top ranking pages down 58% in the 2026 update, and zero click behavior climbing from 53% to 72%. In this video, I break down the exact four step machine Google now uses to turn your best work into one AI answer, why your visibility can keep rising while your traffic falls, and the four assets no platform can ever take from you. This is the same shift my team at NP Digital is helping brands adapt to as search moves from clicks to AI answers. If your strategy still assumes 2024 click behavior, this is where to start. If you'd rather have us adapt your full strategy across SEO, AI search, content, and paid media, that's what we do at

You can rank #1 on Google and still lose more than half your clicks. Most brands haven't noticed yet. For 25 years the internet ran on one deal: you give Google your content, Google sends you traffic. That deal is breaking. Our data shows organic click through rates falling by as much as 61%, top ranking pages down 58% in the 2026 update, and zero click behavior climbing from 53% to 72%. In this video, I break down the exact four step machine Google now uses to turn your best work into one AI answer, why your visibility can keep rising while your traffic falls, and the four assets no platform can ever take from you. This is the same shift my team at NP Digital is helping brands adapt to as search moves from clicks to AI answers. If your strategy still assumes 2024 click behavior, this is where to start. If you'd rather have us adapt your full strategy across SEO, AI search, content, and paid media, that's what we do at

Neil Patel

10,375 görüntüleme • 11 gün önce

The biggest losers in the Google traffic collapse aren't the people who ignored SEO. They're the people who mastered it. For years, businesses built everything around answering questions Google could rank, then Google learned to answer those questions itself. But losing traffic is just the symptom. The real problem is that most companies never owned their audience, they rented it from Google, and the landlord just changed the rules. In this video I break down why the old model is breaking, who gets hit hardest, and why HubSpot could lose most of its organic traffic and still grow revenue. Because clicks and demand were never the same thing, and the brand is the part you actually keep.

Neil Patel

16,348 görüntüleme • 2 ay önce

AI traffic is now surpassing real humans, just as the Dead Internet Theory predicted “Google is killing the internet to benefit its own AI product. For the first time ever, AI bots generated more web traffic than humans this year” That’s happening is when people search something on Google, Google isn’t displaying links first. They are having their AI scan through lots of pages, read the information and then write its own AI summary at the top of the search result This means people creating articles are no longer getting paid with ads by people visiting because they never actually go to the webpage. Instead Google’s AI goes and read the webpages and rewrites them as a summary With this kind of financial loss, there becomes no incentive for people to keep writing articles and websites I found that certain sites are losing up to 97% of their traffic due to this

AI traffic is now surpassing real humans, just as the Dead Internet Theory predicted “Google is killing the internet to benefit its own AI product. For the first time ever, AI bots generated more web traffic than humans this year” That’s happening is when people search something on Google, Google isn’t displaying links first. They are having their AI scan through lots of pages, read the information and then write its own AI summary at the top of the search result This means people creating articles are no longer getting paid with ads by people visiting because they never actually go to the webpage. Instead Google’s AI goes and read the webpages and rewrites them as a summary With this kind of financial loss, there becomes no incentive for people to keep writing articles and websites I found that certain sites are losing up to 97% of their traffic due to this

Wall Street Apes

86,123 görüntüleme • 1 ay önce

AI Overviews are showing up for more and more searches. And when they do, fewer people click. That’s the part everyone is panicking about. Google gives the answer right there, so the user doesn’t always need to visit your site. But SEO is not dead. The game just changed. Because now your job is not only to rank on Google. Your job is to become the source AI trusts enough to recommend. That means showing up in Google AI Overviews, ChatGPT, Perplexity, Gemini, and all the other places people are asking questions. This is why tools like Ubersuggest matter even more now. You can use it to see what people are searching for, what questions they’re asking, what competitors rank for, and where your content gaps are. Then you turn that into better answers. Not generic blog posts. Real expertise. Real examples. Clear explanations. Content that both humans and AI can understand, trust, and cite. And your website still matters because that’s where attention turns into action. AI can summarize information. But it does not close the deal for you. It does not build the relationship. It does not replace your proof, your offer, or your brand. So save this action plan.

Neil Patel

10,260 görüntüleme • 2 ay önce

Blogging used to be simple. Write something helpful, rank on Google, and wake up to traffic. That worked for a long time. Then AI got good at summarizing it. A recent study tracked 100 once-successful blogs. The median one lost 85% of its Google traffic. But the split is what matters. The blogs getting destroyed right now are mostly describing things “best budgeting apps,” “how to start a podcast,” “top tools in 2026.” The ones still growing? They’re proving something. They didn’t write another roundup. They built the thing, tested it for real, spent the money, made the mistake, and showed what actually happened. That’s the new filter. Before you create your next piece of content, look at what people are searching for using then you can create content for what people actually want to know about. Then ask yourself one question: Can AI answer this without me? If the answer is yes, you’re writing something a machine can already summarize better and faster than you. If the answer is no, if it requires your experience, your specific results, your data, or your actual outcome that’s where the opportunity is. The future of content isn’t more articles. It’s proof that can’t be faked.

Blogging used to be simple. Write something helpful, rank on Google, and wake up to traffic. That worked for a long time. Then AI got good at summarizing it. A recent study tracked 100 once-successful blogs. The median one lost 85% of its Google traffic. But the split is what matters. The blogs getting destroyed right now are mostly describing things “best budgeting apps,” “how to start a podcast,” “top tools in 2026.” The ones still growing? They’re proving something. They didn’t write another roundup. They built the thing, tested it for real, spent the money, made the mistake, and showed what actually happened. That’s the new filter. Before you create your next piece of content, look at what people are searching for using then you can create content for what people actually want to know about. Then ask yourself one question: Can AI answer this without me? If the answer is yes, you’re writing something a machine can already summarize better and faster than you. If the answer is no, if it requires your experience, your specific results, your data, or your actual outcome that’s where the opportunity is. The future of content isn’t more articles. It’s proof that can’t be faked.

Neil Patel

11,696 görüntüleme • 2 ay önce

Why did Hitler just get elected? Scott Adams "Some of you are going to wake up and find out that people in your life, let's say young people, who are supporting Kamala Harris, are going to wonder, Why did Hitler just get elected? And you're going to have to answer that question. So rather than answering it yourself, I thought I would do it for you, and then you can just show the video to the young people in your family. How is it that everybody knew he was Hitler. Everybody knew. We all saw it. We saw him trying to steal democracy with our own eyes and stuff. So, so how can he possibly get elected? I give you the answer to that" 1.5x speed

Why did Hitler just get elected? Scott Adams "Some of you are going to wake up and find out that people in your life, let's say young people, who are supporting Kamala Harris, are going to wonder, Why did Hitler just get elected? And you're going to have to answer that question. So rather than answering it yourself, I thought I would do it for you, and then you can just show the video to the young people in your family. How is it that everybody knew he was Hitler. Everybody knew. We all saw it. We saw him trying to steal democracy with our own eyes and stuff. So, so how can he possibly get elected? I give you the answer to that" 1.5x speed

jay plemons

227,827 görüntüleme • 1 yıl önce

What am I watching? Is this real? Why does Trump sound so strange here? Does anyone actually believe that Donald Trump believes any of what he's saying here? In this video Trump apparently says: "You do what God called you to do. You show up. You put one foot in front of the other. You have no idea what God hasssss calling for you. You have no idea where you going, and in some cases you don’t know why. But he needs you. He needs me. He needs all of us. "It is God that is the answer and it is the same turning to God — the ultimate answer to evil — America is a nation strengthened and sustained by God and the prayers of all his children. "Your faith is a force that our enemies can never ever extinguish. Your love of God, family and country is more powerful than any adversaries, hatred, or malice or scorn.”

What am I watching? Is this real? Why does Trump sound so strange here? Does anyone actually believe that Donald Trump believes any of what he's saying here? In this video Trump apparently says: "You do what God called you to do. You show up. You put one foot in front of the other. You have no idea what God hasssss calling for you. You have no idea where you going, and in some cases you don’t know why. But he needs you. He needs me. He needs all of us. "It is God that is the answer and it is the same turning to God — the ultimate answer to evil — America is a nation strengthened and sustained by God and the prayers of all his children. "Your faith is a force that our enemies can never ever extinguish. Your love of God, family and country is more powerful than any adversaries, hatred, or malice or scorn.”

Ed Krassenstein

2,692,434 görüntüleme • 2 yıl önce

Google is making $62 billion a quarter destroying the websites it NEEDS to survive. This is literally a death spiral that ends with Google killing itself. Let me explain what's going on... Google added AI summaries to the top of every search result in 2024. When you Google something now, the answer sits right there on Google's page. You never have to click anywhere. Google took the information from someone else's website, summarized it, and kept you inside Google's ecosystem. The result: 60% of all Google searches now end without a single click to any website. Small publishers lost 60% of their traffic in one year. Medium publishers lost 47%. Even the biggest names in media, the New York Times, the Washington Post, Business Insider, all saw traffic fall between 22% and 55%. The Axios CEO called it "a referral extinction event for the ad-supported web." Google's response to all of this was to tell publishers they can "opt out" of having their content summarized. But opting out also REMOVES your description from normal search results. So the choice Google gives you is let us steal your content for free, or become invisible on the internet. That's extortion. The Washington Post laid off another round of journalists this year because of it. Stereogum, one of the most respected music publications on the internet, had to BEG readers for donations. Business Insider cut 21% of its staff. Dozens of smaller publishers have shut down entirely. The people who actually CREATE the information Google summarizes are going bankrupt while Google posts record revenue. But here's where this gets interesting and where everyone stops thinking: Google's AI summaries are only as good as the content they summarize. If the publishers who write the original articles, run the original investigations, and create the original data go out of business, there is nothing left for Google to summarize. The AI starts recycling old information, the answers get stale, the quality drops, and users start noticing that Google's summaries are increasingly wrong, outdated, or useless. Google is essentially strip-mining the internet for short-term revenue. They are extracting all the value from content creators without paying for it, driving those creators out of business, and then wondering why the quality of their own product is declining. This is exactly what Napster did to the music industry in the early 2000s: Made content free, creators went broke, and quality collapsed. It took a decade to rebuild. Google is doing the same thing to the entire internet at 100x the scale. Rolling Stone, Variety, Deadline, The Hollywood Reporter, and Billboard are now suing Google for antitrust violations. Chegg, the education platform, lost 49% of its traffic and is suing too. The UK's competition authority just ordered Google to let publishers opt out without being punished. The DOJ already ruled Google is an illegal monopoly. And Google's defense in court is genuinely unbelievable. They argue that publishers CHOOSE to let Google index their content and can leave anytime they want. That's like saying you choose to pay protection money to the mob because technically you could close your business and move to another city. Google controls 90% of search. Leaving Google means leaving the internet. Meanwhile Google is investing billions in custom AI chips to make these summaries cheaper at scale. Every quarter the problem gets worse. The internet as we've known it for 25 years ran on a simple deal: Publishers make content. Google sends traffic. Advertisers pay for the traffic. Everyone wins. But Google just BROKE that deal and kept all the money.

Google is making $62 billion a quarter destroying the websites it NEEDS to survive. This is literally a death spiral that ends with Google killing itself. Let me explain what's going on... Google added AI summaries to the top of every search result in 2024. When you Google something now, the answer sits right there on Google's page. You never have to click anywhere. Google took the information from someone else's website, summarized it, and kept you inside Google's ecosystem. The result: 60% of all Google searches now end without a single click to any website. Small publishers lost 60% of their traffic in one year. Medium publishers lost 47%. Even the biggest names in media, the New York Times, the Washington Post, Business Insider, all saw traffic fall between 22% and 55%. The Axios CEO called it "a referral extinction event for the ad-supported web." Google's response to all of this was to tell publishers they can "opt out" of having their content summarized. But opting out also REMOVES your description from normal search results. So the choice Google gives you is let us steal your content for free, or become invisible on the internet. That's extortion. The Washington Post laid off another round of journalists this year because of it. Stereogum, one of the most respected music publications on the internet, had to BEG readers for donations. Business Insider cut 21% of its staff. Dozens of smaller publishers have shut down entirely. The people who actually CREATE the information Google summarizes are going bankrupt while Google posts record revenue. But here's where this gets interesting and where everyone stops thinking: Google's AI summaries are only as good as the content they summarize. If the publishers who write the original articles, run the original investigations, and create the original data go out of business, there is nothing left for Google to summarize. The AI starts recycling old information, the answers get stale, the quality drops, and users start noticing that Google's summaries are increasingly wrong, outdated, or useless. Google is essentially strip-mining the internet for short-term revenue. They are extracting all the value from content creators without paying for it, driving those creators out of business, and then wondering why the quality of their own product is declining. This is exactly what Napster did to the music industry in the early 2000s: Made content free, creators went broke, and quality collapsed. It took a decade to rebuild. Google is doing the same thing to the entire internet at 100x the scale. Rolling Stone, Variety, Deadline, The Hollywood Reporter, and Billboard are now suing Google for antitrust violations. Chegg, the education platform, lost 49% of its traffic and is suing too. The UK's competition authority just ordered Google to let publishers opt out without being punished. The DOJ already ruled Google is an illegal monopoly. And Google's defense in court is genuinely unbelievable. They argue that publishers CHOOSE to let Google index their content and can leave anytime they want. That's like saying you choose to pay protection money to the mob because technically you could close your business and move to another city. Google controls 90% of search. Leaving Google means leaving the internet. Meanwhile Google is investing billions in custom AI chips to make these summaries cheaper at scale. Every quarter the problem gets worse. The internet as we've known it for 25 years ran on a simple deal: Publishers make content. Google sends traffic. Advertisers pay for the traffic. Everyone wins. But Google just BROKE that deal and kept all the money.

Ricardo

251,186 görüntüleme • 4 ay önce

I sat down with Robby Stein (Robby Stein), Google’s VP of Product for Search, at Google I/O. Robby is one of the most interesting product leaders in tech: he helped build Instagram Stories, Reels and Close Friends, and now leads core Google Search products including AI Overviews, AI Mode, Lens and ranking. We talked about one of the biggest shifts in the history of the web: Google Search becoming AI-native. Topics we covered: • AI Mode and whether it is an evolution of Search or a reinvention of it • how Google breaks complex questions into multiple searches behind the scenes • why AI search is much more expensive to run than traditional search • whether Google’s TPUs and infrastructure give it an advantage no one else can match • why Search volume is growing instead of being cannibalized by AI • the tension between great AI answers and traffic for publishers • how Google decides which sources and links to show • what a better internet could look like if AI Search works as intended The big question behind the whole conversation: If Google gives you the answer directly, what happens to the link-based web? A small caveat: sadly the microphones didnt work properly. Therefore the audio quality in this episode isn't perfect due to a recording issue - we appreciate your understanding.

I sat down with Robby Stein (Robby Stein), Google’s VP of Product for Search, at Google I/O. Robby is one of the most interesting product leaders in tech: he helped build Instagram Stories, Reels and Close Friends, and now leads core Google Search products including AI Overviews, AI Mode, Lens and ranking. We talked about one of the biggest shifts in the history of the web: Google Search becoming AI-native. Topics we covered: • AI Mode and whether it is an evolution of Search or a reinvention of it • how Google breaks complex questions into multiple searches behind the scenes • why AI search is much more expensive to run than traditional search • whether Google’s TPUs and infrastructure give it an advantage no one else can match • why Search volume is growing instead of being cannibalized by AI • the tension between great AI answers and traffic for publishers • how Google decides which sources and links to show • what a better internet could look like if AI Search works as intended The big question behind the whole conversation: If Google gives you the answer directly, what happens to the link-based web? A small caveat: sadly the microphones didnt work properly. Therefore the audio quality in this episode isn't perfect due to a recording issue - we appreciate your understanding.

Chubby♨️

30,486 görüntüleme • 4 ay önce

99% of people use AI chatbots wrong. The problem isn't the AI. It's the way most people are prompting it. Matt Wolfe explains exactly why your AI outputs feel average: "AI is the average of the internet. Everything you're going to get is just going to be that middle ground average." "If you steer it, give it your ideas, push it towards the smarter ideas, it's going to help uncover more of that stuff."

99% of people use AI chatbots wrong. The problem isn't the AI. It's the way most people are prompting it. Matt Wolfe explains exactly why your AI outputs feel average: "AI is the average of the internet. Everything you're going to get is just going to be that middle ground average." "If you steer it, give it your ideas, push it towards the smarter ideas, it's going to help uncover more of that stuff."

EllioTrades Clips

12,433 görüntüleme • 4 ay önce

Your website doesn't need a redesign. It needs an honest audit. Most founders spend weeks changing headlines, tweaking CTAs, or redesigning pages... without knowing what's actually stopping people from converting. That's why I like what the team at Kite AI built. At you can get a free Growth Grader that audits your website across messaging, conversions, SEO, AI visibility, pricing, and technical health. It doesn't just tell you what's wrong it prioritizes the fixes that will have the biggest impact and even shows a before/after version of your homepage. The best part? No signup required. Just reply to Kite AI with your website URL, and it'll send back your Growth Grader in about 5 minutes. If you're building a startup, this is probably the easiest way to find out why your website isn't pulling its weight. Check out here:

Your website doesn't need a redesign. It needs an honest audit. Most founders spend weeks changing headlines, tweaking CTAs, or redesigning pages... without knowing what's actually stopping people from converting. That's why I like what the team at Kite AI built. At you can get a free Growth Grader that audits your website across messaging, conversions, SEO, AI visibility, pricing, and technical health. It doesn't just tell you what's wrong it prioritizes the fixes that will have the biggest impact and even shows a before/after version of your homepage. The best part? No signup required. Just reply to Kite AI with your website URL, and it'll send back your Growth Grader in about 5 minutes. If you're building a startup, this is probably the easiest way to find out why your website isn't pulling its weight. Check out here:

Shruti Codes

50,380 görüntüleme • 2 ay önce

WATCH THIS 🔥🤯 Martha MacCallum just asked the question Democrats apparently CAN’T answer: Why is it “wrong” to send every adult American a $5,000 check… …but perfectly fine for the government to take your tax dollars and spend them however Washington wants?? Rep. Jason Crow says the $5,000 “Trump Dividend” is wrong. So Dems would rather keep control of YOUR money than let you spend it on your own family?? How is that not insane?? 💸🚨 “Here’s a check. You spend it the way you see fit.” That’s the whole argument. And they still said no. How crazy is that?

WATCH THIS 🔥🤯 Martha MacCallum just asked the question Democrats apparently CAN’T answer: Why is it “wrong” to send every adult American a $5,000 check… …but perfectly fine for the government to take your tax dollars and spend them however Washington wants?? Rep. Jason Crow says the $5,000 “Trump Dividend” is wrong. So Dems would rather keep control of YOUR money than let you spend it on your own family?? How is that not insane?? 💸🚨 “Here’s a check. You spend it the way you see fit.” That’s the whole argument. And they still said no. How crazy is that?

CONSTITUTION X 🇺🇸

53,190 görüntüleme • 18 gün önce

Why nobody is talking about this loud? West is stealing your Yoga and selling it back to you and they own the patents now. They sold the stigma to Indians and looted the business derived out of Yoga.

Why nobody is talking about this loud? West is stealing your Yoga and selling it back to you and they own the patents now. They sold the stigma to Indians and looted the business derived out of Yoga.

Pattern Preacher

26,420 görüntüleme • 3 ay önce

Coinbase CEO Explains “Reverse Prompting” and the Rise of the AI CEO Brian Armstrong: “One of the big pushes we made in the last year was we got our own internal hosted AI model that was connected to all of our data sources, right?” “So it's like every Slack message, every Google doc, Salesforce data, Confluence, you know.” “So now the data is all aggregated and I've started to ask it really… it's not just like prompting it, ‘Hey, can you write this kind of memo for me,’ or something.” “I'm asking these AI agents now, ‘As CEO, what should I be aware of in the company that I might not be aware of?’ And it'll tell me, ‘Did you know that there's actually disagreement on this team about the strategy?’ And I was like, actually, I didn't know that.” “This is like reverse prompting. So instead of telling the AI agent what you want it to do, you ask it what you should be thinking more about.” @jason: “It's a mentor. It's a coach.” Brian: “Yeah. Like, what could make me a better CEO? And it's like, ‘Well, I looked at how you spent your time in the last quarter and here's how you said that you wanted to spend it, but you actually spent 32% of your time on this instead of 20%.’” “I've asked it other questions like, ‘What's the thing that I changed my mind on the most over the last year?’ Things like that.” “It'll prompt you with information you should be thinking about instead of the other way around.” Thanks to our partner for making this happen!: Our episode is sponsored by the New York Stock Exchange - a modern marketplace and exchange for building the future. It all happens at the NYSE 🏛.

Coinbase CEO Explains “Reverse Prompting” and the Rise of the AI CEO Brian Armstrong: “One of the big pushes we made in the last year was we got our own internal hosted AI model that was connected to all of our data sources, right?” “So it's like every Slack message, every Google doc, Salesforce data, Confluence, you know.” “So now the data is all aggregated and I've started to ask it really… it's not just like prompting it, ‘Hey, can you write this kind of memo for me,’ or something.” “I'm asking these AI agents now, ‘As CEO, what should I be aware of in the company that I might not be aware of?’ And it'll tell me, ‘Did you know that there's actually disagreement on this team about the strategy?’ And I was like, actually, I didn't know that.” “This is like reverse prompting. So instead of telling the AI agent what you want it to do, you ask it what you should be thinking more about.” @jason: “It's a mentor. It's a coach.” Brian: “Yeah. Like, what could make me a better CEO? And it's like, ‘Well, I looked at how you spent your time in the last quarter and here's how you said that you wanted to spend it, but you actually spent 32% of your time on this instead of 20%.’” “I've asked it other questions like, ‘What's the thing that I changed my mind on the most over the last year?’ Things like that.” “It'll prompt you with information you should be thinking about instead of the other way around.” Thanks to our partner for making this happen!: Our episode is sponsored by the New York Stock Exchange - a modern marketplace and exchange for building the future. It all happens at the NYSE 🏛.

The All-In Podcast

80,524 görüntüleme • 8 ay önce

Be the Answer. Rank Across Google & AI. "SEOitis" is live SEO is no longer just about ranking on Google. ChatGPT, Claude, Gemini, Perplexity, and Grok are answering your customers questions - with or without you. We built SEOitis to fix that. > Track how AI engines talk about your brand > Find the gaps > Research, write, and publish - one pipeline 🔗

Kalash

87,866 görüntüleme • 3 ay önce

Boris Cherny, the creator of Claude Code at Anthropic, just listed 9 patterns that waste 73% of your tokens. in this podcast he breaks down exactly how the model burns tokens before it even reads your prompt: - the 14% you lose to CLAUDE.md before typing a word - the 13% you pay re-reading old chat history - the 11% from hooks you forgot you installed - why most "Claude got dumber" complaints are wrong if you're hitting Max limits more than once a week, you have at least 4 of these. Probably 7. instead of another show tonight, watch this. my own breakdown based on 400+ hours of usage is below, read it after the podcast

Boris Cherny, the creator of Claude Code at Anthropic, just listed 9 patterns that waste 73% of your tokens. in this podcast he breaks down exactly how the model burns tokens before it even reads your prompt: - the 14% you lose to CLAUDE.md before typing a word - the 13% you pay re-reading old chat history - the 11% from hooks you forgot you installed - why most "Claude got dumber" complaints are wrong if you're hitting Max limits more than once a week, you have at least 4 of these. Probably 7. instead of another show tonight, watch this. my own breakdown based on 400+ hours of usage is below, read it after the podcast

Mnimiy

1,383,148 görüntüleme • 5 ay önce

you read it and pictured yourself approving instead of watching approving what five agents, each on its own machine, handing work down the line, and what reaches you at the end is the last one's account of the four accounts before it the honest part of that piece says you do not own the harness that is the exact layer where a check on any of this would have to live a hundred and twenty a seat, three hundred at the top, about six times what you pay now you are not buying the work you are buying a shorter story about the work, told by the only party that watched it happen

you read it and pictured yourself approving instead of watching approving what five agents, each on its own machine, handing work down the line, and what reaches you at the end is the last one's account of the four accounts before it the honest part of that piece says you do not own the harness that is the exact layer where a check on any of this would have to live a hundred and twenty a seat, three hundred at the top, about six times what you pay now you are not buying the work you are buying a shorter story about the work, told by the only party that watched it happen

Lagosaur

23,498 görüntüleme • 1 ay önce

Former Meta Chief AI Scientist Yann LeCun on the three paradigms of machine learning — and why the third is what made ChatGPT possible: Here's each one, and where it breaks. First, supervised learning. You tell the machine the answer. "You show it a picture, let's say of a table, and you tell it this is a table. So it's supervised because you tell it what the correct answer is." Get it wrong, and the machine rewrites itself: "The system computes its output, and if it says something else than table, then it's going to adjust its parameters, its internal structure, so that the output it produces gets closer to the output you want." Repeat at scale and something more than memorisation appears: "Eventually the system will find a way to recognize every image you trained it on, but also images it's never seen that are similar to the one you train it on. This is called a generalization ability." The limit: a human has to supply every single answer. That doesn't scale to the size of the internet. Second, reinforcement learning. You don't give the answer, only a verdict. "You don't tell the system what the correct answer is. You only tell it whether the answer it produced was good or bad." Learning to ride a bike, essentially: "You try to ride a bike and you don't know how to ride the bike and after a while you fall. So you know you did something bad and so you change your strategy a little bit. And eventually you learn how to ride a bike." For years the field assumed this was the closest thing to how animals actually learn. Yann LeCun's verdict: "Now it turns out reinforcement learning is extremely inefficient." It dominates wherever failure is free: "It works really well if you want to train a system to play chess or play go or poker, because you can have the system play millions and millions of games against itself and basically fine-tune itself. But it doesn't really work in the real world." The limit, in one image: "If you want to train a car to drive itself, you're not going to do it with reinforcement learning. It's going to crash thousands of times." On robotics he's careful rather than dismissive: "Reinforcement learning can be part of the solution, but it's not the complete answer. It's not sufficient." Third, self-supervised learning. You tell the machine nothing at all. "And this is what has enabled the recent progress in natural language understanding and chatbots." The strange part is that you stop asking for a task: "You don't train the system to accomplish any particular task. You just train it to basically capture the structure..." The method is deliberate sabotage: "You take a piece of text, you corrupt it in some way, by for example removing some words, and then you train a big neural net to predict the words that are missing." And one narrow version of that trick runs every chatbot on Earth: "A special case of this is that you take a piece of text and the last word in that text is not visible, and so you train the system to predict the last word in that text — and this is the way large language models are trained on." So why did the third one win? Supervised learning needs a human. Reinforcement learning needs a crash. Self-supervised learning needs neither — because the missing word and the correct answer are the same thing. The data grades itself.

Former Meta Chief AI Scientist Yann LeCun on the three paradigms of machine learning — and why the third is what made ChatGPT possible: Here's each one, and where it breaks. First, supervised learning. You tell the machine the answer. "You show it a picture, let's say of a table, and you tell it this is a table. So it's supervised because you tell it what the correct answer is." Get it wrong, and the machine rewrites itself: "The system computes its output, and if it says something else than table, then it's going to adjust its parameters, its internal structure, so that the output it produces gets closer to the output you want." Repeat at scale and something more than memorisation appears: "Eventually the system will find a way to recognize every image you trained it on, but also images it's never seen that are similar to the one you train it on. This is called a generalization ability." The limit: a human has to supply every single answer. That doesn't scale to the size of the internet. Second, reinforcement learning. You don't give the answer, only a verdict. "You don't tell the system what the correct answer is. You only tell it whether the answer it produced was good or bad." Learning to ride a bike, essentially: "You try to ride a bike and you don't know how to ride the bike and after a while you fall. So you know you did something bad and so you change your strategy a little bit. And eventually you learn how to ride a bike." For years the field assumed this was the closest thing to how animals actually learn. Yann LeCun's verdict: "Now it turns out reinforcement learning is extremely inefficient." It dominates wherever failure is free: "It works really well if you want to train a system to play chess or play go or poker, because you can have the system play millions and millions of games against itself and basically fine-tune itself. But it doesn't really work in the real world." The limit, in one image: "If you want to train a car to drive itself, you're not going to do it with reinforcement learning. It's going to crash thousands of times." On robotics he's careful rather than dismissive: "Reinforcement learning can be part of the solution, but it's not the complete answer. It's not sufficient." Third, self-supervised learning. You tell the machine nothing at all. "And this is what has enabled the recent progress in natural language understanding and chatbots." The strange part is that you stop asking for a task: "You don't train the system to accomplish any particular task. You just train it to basically capture the structure..." The method is deliberate sabotage: "You take a piece of text, you corrupt it in some way, by for example removing some words, and then you train a big neural net to predict the words that are missing." And one narrow version of that trick runs every chatbot on Earth: "A special case of this is that you take a piece of text and the last word in that text is not visible, and so you train the system to predict the last word in that text — and this is the way large language models are trained on." So why did the third one win? Supervised learning needs a human. Reinforcement learning needs a crash. Self-supervised learning needs neither — because the missing word and the correct answer are the same thing. The data grades itself.

Big Brain AI

49,293 görüntüleme • 1 ay önce

Google’s golden age is coming to an end. For the last 20 years, it has dominated the search engine world, but the tides are shifting. User behavior is evolving, competitors are rising, and Google’s reliance on ads is driving people away. SEO as we know it is about to change—forever. In this video, I’ll break down exactly why Google is dying, what it means for the future of search, and how you can adapt your SEO strategy to stay ahead. We’ll dive into Google’s rise, its slow decline, and the new search landscape emerging in its place. If you want to future-proof your SEO strategy, this is a must-watch.

Google’s golden age is coming to an end. For the last 20 years, it has dominated the search engine world, but the tides are shifting. User behavior is evolving, competitors are rising, and Google’s reliance on ads is driving people away. SEO as we know it is about to change—forever. In this video, I’ll break down exactly why Google is dying, what it means for the future of search, and how you can adapt your SEO strategy to stay ahead. We’ll dive into Google’s rise, its slow decline, and the new search landscape emerging in its place. If you want to future-proof your SEO strategy, this is a must-watch.

Neil Patel

17,621 görüntüleme • 1 yıl önce