Luyện nói tiếng Anh bằng Shadowing qua video: How Apple Accidentally Won The Al Race

The AI race is the largest single year corporate investment in the history of business, and every dollar of it is a bet that whoever builds the biggest brain wins.

This year, Microsoft, Amazon, Google, and Meta will spend over $725 billion building data centers for artificial intelligence.

But in March 2026, while hundreds of billions of dollars was being spent on data centers and model development,

Apple held a quiet launch event in New York and announced a $599 laptop that runs on an iPhone chip.

And you might have noticed that the person Apple sent out to reveal it wasn't Tim Cook.

It was a man named John Ternus.

Just weeks later, Apple named Ternus as its next CEO, taking over when Cook retires later this fall.

So on September 1st, John Ternus, a hardware engineer almost nobody outside of Apple or the tech sphere can name, takes over the most valuable company in the world in the middle of an AI race

that most people say Apple is dead-lasting.

But this January, Apple did something that nobody saw coming.

It agreed to pay its biggest competitor a billion dollars a year to rent the AI models

that run in its own products.

Why would the most valuable company on earth hand the most

personal part of its products to the one rival it has spent a decade trying to escape?

To answer that, we have to go back in time, because the last time Apple wrote a check like this to a rival, the year was 1996, the company was 12 weeks away from running out of cash,

and the check it wrote bought them the next 25 years.

But before I tell you what I think is happening, I want to show you what I found.

I went back through Apple's entire paper trail, the court settlements, the filings, the deal terms, the org chart, and buried in all of it,

I mapped out what I believe is Apple's five-phase plan to win the AI race.

But for this to make sense, I need to show you how bad everything got for Apple, starting with a mistake that cost Apple a quarter of a billion dollars.

Phase one, $250 million mistake.

So let me make the case against Apple first because it's a good one.

In 2023, at its developer conference, Apple stood on stage and promised a new generation of Siri.

They said that it could see what was on your screen, understand the context of your life, and take actions on your behalf across all apps.

It was going to be the assistant that Siri was always supposed to be.

And Apple sold the iPhone 16 based around that entire promise.

Then, in the spring of 2025, two years after the announcement of the new Siri, Apple quietly delayed those features and pulled the ads.

The product sold marketing campaign leaned on features that simply didn't even exist yet.

And this year, it paid for it.

Apple agreed to a $250 million settlement over false advertising, covering roughly 37 million iPhones,

which is somewhere between $25 and $95 per device.

But then in January 2026, everything changed.

Apple announced that the next generation of its own AI models would be built on Google's Gemini.

By Bloomberg's estimate, the deal runs around $1 billion per year.

Now, if you stop the story right there, this looks like a company that gave up competing in the AI race.

But I want to show you what everyone is missing when covering this story.

Because almost everything I just described has happened to Apple before.

The last time it happened, Apple was 12 weeks from running out of cash and declaring bankruptcy.

But what it did next would go on to become the biggest comeback story in corporate history.

Phase 2.

The time Apple almost went bankrupt.

Let's go back to 1996, where Apple was far from being the multi-trillion dollar company it is today.

In the summer of 1996, Apple was 12 weeks away from bankruptcy.

It has spent years trying to build a modern operating system for their computers called Copland, failing for years.

The one piece of software Apple depended on to ship its hardware, it couldn't build.

The industry experts had already written Apple's obituary.

Michael Dell, the founder of Dell Computers, was asked on stage what he would do if he were running Apple.

His answer became infamous in the tech world.

Dell said that if he were running Apple, he would shut it down and give the money back to the shareholders.

The smartest people in the tech industry looked at Apple and saw a dead company.

And it made things even stranger.

The main reason Apple wasn't able to develop a modern operating

system in the first place traces back to a decision Apple made over a decade earlier. In 1985,

Apple's own board pushed Steve Jobs out of the very company he co-founded after a power struggle with the new CEO, John Scully.

So Jobs left and he started a new company called Next.

Spending the next 10 years building exactly the kind of advanced operating system Apple was now failing to build on its own.

So when Apple went looking for software to save itself in 1996, the best option on the table belonged to the founder that it had thrown out.

So Apple paid around 429 million dollars for Next and brought Steve Jobs back.

That software from Next became the groundwork for Mac OS X, which became the foundation for the software on the original iPhone

and then became the software in every product Apple has sold since.

A company 12 weeks from death bought the one layer it couldn't build, wrapped in its own hardware, and wrote it for a quarter of a century.

Does any of this sound familiar yet?

Oh, and then a year later, Apple made a second unthinkable decision.

In August of 1997, Steve Jobs walked out on stage at Macworld in Boston and announced that Microsoft,

the great enemy, the company Mac users had been trained to hate, was investing $150 million into Apple.

Funny enough, when Bill Gates appeared on a giant screen behind Jobs, the crowd booed.

They actually booed.

But this was the beginning of a generational come back.

At that time, Apple was losing about a billion dollars a year and had less than 90 days of cash left.

Microsoft's investment bought Apple the runway it needed to survive just long enough to build the iMac

and then the iPod and then everything else that would come next.

The iMac landed in 1998, the iPod in 2001.

Apple was profitable again the very next year.

The Google Gemini deal today is both of those moves at the same time.

Apple can't build its own AI model fast enough, so it's renting the layer and it's taking that layer from the rival that's already done all of the hard work.

This is the oldest move in the Apple playbook, a move that Apple runs again and again.

In April 2008, Apple bought a tiny chip company called PA Semi, and that became Apple's own silicon chips, the chips that now beat Intel and ship on every Apple device.

In 2014, Apple bought Beats for around $3 billion, and everybody only talked about the headphones, but almost nobody paid attention to the other thing Apple got in that Beats acquisition,

a small streaming service called Beats Music, which most people just wrote off as a throw-in.

Apple shut Beats Music down, rebuilt it, and relaunched it as Apple Music, which is now estimated to have 100 million paying subscribers.

Apple Music rolls up into Apple's service business, which just posted a record of about $31 billion in a single quarter.

The shape is always the same.

You know, Apple buys the layer it can't build fast enough, wraps it into their own hardware and design, absorbs it into their ecosystem, and keeps the customer.

Real quick, if you've made it this far, you're probably the kind of person who wants to see where technology is headed before it gets there.

It's exactly what this channel was built for.

We also sit down with the people actually building the future and ask the questions you would want to ask them.

I've got some insane guests booked over the next few months

with people you've definitely heard of building the things you're probably already using every day.

That sounds like your kind of thing.

Hit subscribe so you don't miss it.

All right, now let's dive into the piece of Apple's strategy that none of its competitors can copy.

Phase 3.

Why the world's smartest AI model can't beat Siri.

Fast forward to June 2026.

At his final developer conference's CEO, Tim Cook walked out and unveiled the thing Apple had been failing to ship for two years.

A completely rebuilt Siri.

Apple even gave it a new name.

It's now called Siri AI and it has its own standalone app.

Underneath, Siri AI runs on a custom Gemini model in Apple's cloud paired with Apple's own smaller models on the device.

100

And it can now finally do the things Apple promised back in 2024.

101

Siri AI can see what's on your screen, pull contacts from your emails and your messages and your photos, and carry out actions across your apps.

102

Now, let me be honest about where it still stands, because the Frontier chatbots are still smarter than Siri.

103

They write better, they reason better, they code better.

104

I mean, if raw intelligence were the whole game, Apple would have already lost it.

105

But here's what those chatbots don't have.

106

Intelligence and access are two completely different advantages.

107

The smartest model in the world is useless to you if it has no context of your work and your personal life.

108

The reason why that rebuilt Siri matters so much is the orchestrator underneath it.

109

The part that decides what to answer on the device itself

110

and what to hand up to Apple's private cloud to process to process.

111

That is what the Google deal actually bought.

112

For ChadCBT or Gemini or Claude to match this on your iPhone, you would have to hand a third-party standing access to your messages, your mail, your photos, your calendar, and your screen.

113

Every one of those is a trust decision, and I think most people will say no. Now, this level of trust with its consumers didn't happen overnight for Apple.

114

I mean, they've spent more than a decade building trust with their audience in a way

115

that almost every other tech company has failed to do so.

116

After the San Bernardino attack, the FBI got a federal court order demanding that Apple build a special version of iOS to break into the shooter's iPhone.

117

Apple said no. Tim Cook later published an open letter refusing a court order in a terrorism case

118

because building that tool one time would have put a back door into every iPhone on Earth.

119

Think about how rare that is.

120

A company told the FBI no on privacy and made it stick.

121

That reputation is the ground that Apple's trust moat is built on.

122

Today, the major AI labs have exactly one way around that moat.

123

Get you off of the iPhone and Apple ecosystem entirely. And

124

that quickly turns the whole AI race into a hardware war

125

on the one battlefield where Apple has won for 15 years straight.

126

The model is becoming a commodity.

127

I mean, several companies now have one that is good enough.

128

The device, the personal context, and the distribution across a couple of billion phones, those are not commodities.

129

There's a problem with all this.

130

The moat cuts both ways.

131

If the AI model stops being a commodity, then Apple has just made itself dependent on a company you spent 10 years positioning against.

132

I'm not gonna pretend that there's an easy answer that resolves that risk cleanly.

133

There isn't an easy answer, and it's the single biggest risk in Apple's AI strategy.

134

But sitting underneath that risk is a hidden bet

135

that the rest of the industry can't match because Apple has already built a computing network worth trillions, and it is running in your pocket right now.

136

Phase four, the hidden trillion dollar AI computing network.

137

Here's the hidden bet underneath Apple's AI bet.

138

Every other company in the AI race has to build the computer that runs the AI itself.

139

Remember that $725 billion number that I mentioned at the start of this video?

140

That's the cost of building those computers.

141

Giant data centers full of chips, burning power, and running somebody else's model in the cloud.

142

Apple already shipped its AI computer.

143

Apple has sold well over 2 billion iPhones since the first generation in 2007, and around 1.5 billion of them are still being used today.

144

Then you add the iPads, the Macs, the watches, and the rest of its products, and Apple has roughly 2.5 billion active devices out in the world, each one with an Apple chip inside of it.

145

Apple's customers already built the infrastructure, one device at a time.

146

Apple would rather pay Google a billion dollars a year to run Gemini inside its own data centers

147

and then shrink it down to run on your phone than chip your personal life out to somebody else's cloud.

148

The whole AI industry is spending itself living in the cloud, but Apple is quietly betting that the future runs locally on the device.

149

And the economics of that run brutally in Apple's favor.

150

Every time a cloud chatbot answers a question, it costs the company that runs it real money in compute and in power.

151

Every time the new Siri answers on the device, after the iPhone is already sold, it costs Apple close to nothing.

152

I mean, the rest of the industry is building a business where every use is a new cost.

153

Apple is building one where the customer already bought the hardware that does the majority of the work.

154

And this is the wall the AI labs are about to run into.

155

Say, OpenAI wants to escape Apple's moat by building its own device.

156

Spoiler alert, they're already trying to.

157

Any one new product that OpenAI or any other company launches does so into a world where Apple already ships an iPhone, an iPad, a MacBook, an iMac,

158

an Apple Watch, AirPods, and an Apple TV, and every one of them is built to hand off to the next.

159

Your techs start on the phone and then finishes on the laptop.

160

Your photos are on all of them at once.

161

That interlocking system took Apple almost 20 years and billions of devices to build.

162

Matching it on a first launch with a single gadget is close to impossible.

163

Even Johnny Ive is reportedly building something meant to work with your iPhone rather than replace it.

164

And this is where the founder's vision comes all the way back around.

165

The Steve Jobs vision was always about total product integration.

166

Own the silicon, own the system, own the device.

167

Renting the model layer for a few years while Apple figures the rest out is how Apple protects that.

168

On the same April day it named Turnus, the company's new CEO, Apple also created a job title that didn't exist the day before.

169

Chief Hardware Officer.

170

They appointed the role to Johnny Sruji, the engineer behind every Apple chip since 2008

171

and folded all of Apple's hardware groups into the one organization reporting directly up to Ternus.

172

Hardware leads the company's direction now.

173

Jobs built the products, Cook built the machine that could manufacture them at a scale of the entire planet, and Ternus will point that machine back at hardware.

174

But the real story of John Ternus' CEO takeover is stranger than the announcement itself, because the next CEO of Apple had been hiding in plain sight for six years.

175

Phase five, the CEO hiding in plain sight.

176

In 2020, Tim Cook hands Turnus control of iPhone hardware, the crown jewels of the entire company.

177

In 2021, Turnus takes over all of Apple's hardware engineering and gets his first seat at the executive table.

178

In 2022, Turnus picks up the watch, then robotics, then the design team.

179

And then, on March 4th, 2026, Turnus fronts the entire MacBook Neo launch.

180

He kicks off the event, and he does the whole media run that's normally the chief executive's job.

181

Seven years later, Apple announced John Turnus would be replacing Cook as CEO starting September 1st, 2026.

182

And one detail gives the planned timeline of the CEO succession away completely.

183

Back in June, the analyst Ming-Chi Kuo reported

184

that Apple's entire headset roadmap had been cut down from seven planned products down to only two, and that the decision was signed off by John Turnus.

185

Three months before it was announced that he had the CEO role, Turnus was already making the biggest calls at Apple that would set their direction up for the next decade.

186

So who exactly is this John Turnis guy anyway?

187

Back in the 1990s, Turnis' senior engineering project in college was a robotic feeding arm for people with paralysis.

188

His first job out of school was at a virtual reality headset company that ultimately didn't make it.

189

He lived through the first failed VR wave firsthand, which is exactly why, years later, he was one of the biggest skeptics at Apple about the Vision Pro.

190

And he turned out to be right.

191

The product went on to completely underperform expectations.

192

Ternus is the one who pushed to split the iPad software on its own track from iOS

193

because the hardware had raced out ahead of the software of the iPhone.

194

But let me be fair because Ternus does have a losing column at Apple too.

195

He championed the MacBook Touch Bar, the strip of screen nobody has to keep, and the faulty butterfly keyboard shipped on his watch.

196

And Apple settled a class action over it for $50 million.

197

Some people inside Apple say Ternus is better at maintaining products than inventing them.

198

The man who just inherited the Google Bridge is the same man who built the hardware

199

that bridge is supposed to lead to.

200

So let me come back to the question I asked at the start.

201

Why would the most valuable company on earth hand the most

202

personal part of its products to the rival who spent a decade trying to escape?

203

Because Apple has done this exact same thing twice before.

204

And both times it was a setup for a comeback nobody saw coming.

205

In 1996, Apple bought the software it couldn't build itself from the founder it had fired.

206

In 97, it took the money and the lifeline from the rival everyone feared, and it used both to buy the runway to build the

207

products it actually wanted on its own hardware under its own control.

208

Apple's Google deal is that same move.

209

History has a habit of repeating itself, and Apple is betting everything that it will win the AI race.

210

Make sure to subscribe, and I'll see you on the next Frontier.

Phương Pháp Shadowing Là Gì?

Shadowing là kỹ thuật học ngôn ngữ có cơ sở khoa học, ban đầu được phát triển cho chương trình đào tạo phiên dịch viên chuyên nghiệp và được phổ biến rộng rãi bởi nhà đa ngôn ngữ học Dr. Alexander Arguelles. Nguyên lý cốt lõi đơn giản nhưng cực kỳ hiệu quả: bạn nghe tiếng Anh của người bản xứ và lặp lại to ngay lập tức — như một "cái bóng" (shadow) đuổi theo người nói với độ trễ chỉ 1–2 giây. Khác với luyện ngữ pháp hay học từ vựng bị động, Shadowing buộc não bộ và cơ miệng phải đồng thời xử lý và tái tạo ngôn ngữ thực tế. Các nghiên cứu khoa học xác nhận phương pháp này cải thiện đáng kể phát âm, ngữ điệu, nhịp điệu, nối âm, kỹ năng nghe và độ lưu loát khi nói — đặc biệt hiệu quả cho người luyện IELTS Speaking và muốn giao tiếp tiếng Anh tự nhiên như người bản ngữ.

Luyện nói tiếng Anh bằng Shadowing qua video: How Apple Accidentally Won The Al Race

Bạn sẽ học được gì

Nghe để nhận diện những âm thanh này

Nói như người bản ngữ

Phương Pháp Shadowing Là Gì?