Post

How to get URL link on X (Twitter) App

On the Twitter thread, click on or icon on the bottom
Click again on or Share Via icon
Click on Copy Link to Tweet
Paste it above and click "Unroll Thread"!
More info at Twitter Help

Varun Navani

@VarunNavani

Jan 8 • 18 tweets • 6 min read • Read on X

Scrolly

In 2017, Tesla faced an impossible challenge:

Teaching cars to see and understand the world like humans.

Then they hired a 29-year-old genius who solved it.

But what he did next shocked everyone in Silicon Valley.

Here's the untold story of Tesla's AI mastermind:

Imagine teaching a computer to understand the world like a human.

Not just recognizing objects, but understanding context, predicting movement, and making split-second decisions.

This was Tesla's challenge in 2017. Their autonomous driving program needed advancement.

But they knew something others didn't...

Traditional car companies were using a combination of sensors:

• LIDAR (laser-based detection)
• Radar systems
• Ultrasonic sensors
• Multiple cameras

But Tesla wanted something different.

And their unconventional approach would change everything...

The problem seemed impossible:

Most experts believed you couldn't rely on cameras alone.

The human brain processes visual information in complex ways that computers hadn't yet replicated.

Then they discovered a hidden genius...

Born in Slovakia, Andrej Karpathy was a computer science prodigy.

By 29, he'd already:

• Completed his PhD at Stanford
• Become a founding member of OpenAI
• Specialized in computer vision and deep learning

But his most revolutionary idea was yet to come...

Karpathy's vision was distinct:

Instead of relying on multiple sensor types, he believed cars could learn to "see" using primarily cameras.

Just like humans don't need radar to drive, cars shouldn't either.

This insight would transform the industry forever...

The core of Tesla's vision system was the HydraNet architecture:

A neural network that could process data from 8 cameras simultaneously.

It could understand depth, detect objects, and make decisions in real time.

But there was one massive obstacle standing in their way:

To work properly, the system needed enormous amounts of training data.

Traditional companies spent years manually collecting and labeling data.

Karpathy implemented the Data Engine:

And what happened next was unprecedented...

Every Tesla on the road became part of the data collection network:

• Recording driving scenarios
• Learning from human responses
• Identifying edge cases
• Improving through real-world experience

The results would shock everyone...

Tesla's vision system could now:

• Process multiple tasks simultaneously
• Detect objects and predict movement
• Make rapid driving decisions

The automotive world took notice.

But then something unexpected happened...

In July 2022, at the height of Tesla's AI development, he left the company.

The man who revolutionized their autonomous driving approach stepped away.

He saw broader applications for AI beyond automotive:

This decision would reshape the future of AI...

Karpathy believed AI would transform multiple industries:

• Healthcare
• Education
• Scientific research
• Creative work

Cars were just the beginning.

And he was about to prove it...

His departure was significant news in Silicon Valley.

His legacy at Tesla includes:

• A vision-based approach to autonomy
• A sophisticated data collection system
• Advanced neural network architecture

But the real breakthrough was still coming...

Today, Karpathy's vision continues to shape AI development:

Not just in autonomous vehicles...

But across every industry that needs to understand and process visual information.

And this reveals something crucial about AI adoption:

The biggest breakthroughs don't come from adding more sensors or complexity.

They come from reimagining what's possible with existing technology.

Just like Tesla revolutionized autonomous driving with cameras...

Your business can transform with the right AI approach.

This is why I help businesses and institutions navigate AI adoption.

Not by adding unnecessary complexity...

But by finding the simplest, most effective solutions.

Because sometimes, less really is more.

https://twitter.com/228418683/status/1877037740484886870

I hope you've found this thread helpful.

Follow me @VarunNavani for more.

Like/Repost the quote below if you can:

https://twitter.com/228418683/status/1877037740484886870

Video credits:
• YT link youtube.com/watch?v=W-ubNv…
• YT link youtube.com/watch?v=xxOOpW…
• YT link youtube.com/watch?v=cdiD-9…

• • •

Missing some Tweet in this thread? You can try to force a refresh

This Thread may be Removed Anytime!

Twitter may remove this content at anytime! Save it as PDF for later use!

More from @VarunNavani

Varun Navani

@VarunNavani

Feb 28

The most misunderstood fact about AI:

Everyone thinks it's becoming intelligent.

What scientists just discovered proves every tech company has been lying about these systems:

AI isn't becoming intelligent at all.

In fact, it's not even close to what we think it is.

The term "artificial intelligence" itself contributes to misconceptions by "anthropomorphizing" what is essentially a sophisticated computational tool.

But what does this MEAN?

Here's what's actually happening:

When you talk to ChatGPT, it's not "thinking" or "understanding" you.

It operates solely on pattern recognition and probability calculations through algorithms trained on specific datasets.

But there's an even bigger deception at play:

Read 16 tweets

Varun Navani

@VarunNavani

Feb 24

The AI battle that matters is NOT Musk vs Altman.

It's between two childhood friends turned bitter rivals.

What started in a London casino ended in "betrayal".

And their different visions will shape humanity's future🧵

Meet Demis Hassabis and Mustafa Suleyman.

One a chess prodigy who studied at Cambridge.

The other, the son of a Syrian taxi driver who dropped out of Oxford.

Their paths first crossed in North London in the 1990s, when Suleyman befriended Hassabis's younger brother...

Despite an 8-year age gap, they shared a burning vision:

To build artificial intelligence that could change the world.

In 2010, they reunited at London's Victoria Casino.

Over poker games, they dreamed up what would become DeepMind:

Read 15 tweets

Varun Navani

@VarunNavani

Feb 19

xAI just made the greatest discovery about human learning.

Their $4B supercomputer in Memphis analyzed millions of data points.

In 122 days, they found something extraordinary about how our minds work.

They revealed why we've been learning wrong our entire lives:

It started with a massive investment.

xAI built Colossus - a supercomputer with 200,000 NVIDIA GPUs in Memphis, Tennessee.

Cost: $4 billion.

But what they discovered about learning would change education forever...

The first revelation came during training:

Grok 3 wasn't just learning - it was learning HOW to learn.

Through synthetic data and advanced reasoning, it developed the ability to master new subjects at unprecedented speeds.

But here's where it gets fascinating:

Read 14 tweets

Varun Navani

@VarunNavani

Feb 17

Elon Musk's required reading list revealed:

• Not just business books
• Not just tech manuals
• But warnings about humanity's future

Here's what he wants everyone to understand:

At age 9, Musk had already read the entire Encyclopedia Britannica.

His teenage years were filled with 10 hours of daily reading - a habit that would shape the future of multiple industries.

This obsessive pursuit of knowledge had a deeper purpose that would soon become clear:

The Hitchhiker's Guide to the Galaxy became his philosophical foundation.

The book taught him a profound lesson: asking the right questions matters more than having all the answers.

This insight would later transform how he approached complex technological challenges.

His intellectual journey was about to take a fascinating turn:

Read 19 tweets

Varun Navani

@VarunNavani

Feb 12

NEOM: The greatest urban experiment in history.

A city with no cars, no roads, and zero emissions.

Powered entirely by AI and renewable energy.

Here's what living there will actually look like:

Imagine waking up in a 170km-long mirrored skyscraper.

Outside your window, a perfect climate year-round.

No traffic. No pollution. No chaos.

But this isn't science fiction. Construction has already begun...

Saudi Arabia is building NEOM:

A $500 billion mega-city spanning 26,500 km².

It's larger than some countries and powered by 100% renewable energy.

And it's redefining what's possible in urban living.

Here's where things get interesting:

Read 17 tweets

Varun Navani

@VarunNavani

Feb 7

She exposed the truth about AI bias.

Google gave her an ultimatum: Retract or leave.

She chose integrity.

Here's how she exposed AI's biggest problem:

In 2020, Gebru was co-lead of Google's Ethical AI team.

Her groundbreaking research exposed major biases in facial recognition systems, showing error rates of 34.7% for dark-skinned females vs 0.8% for light-skinned males.

But when she questioned Google's AI development practices, something unprecedented happened...

She co-authored a paper examining the hidden dangers of large language models.

The findings were alarming:

Training a single AI model could emit as much carbon as 5 cars' lifetime emissions.

But Google's response left the tech world stunned: