NVIDIA CEO Jensen Huang's Vision for the Future

From One Thing at a Time to Many: How Computers Got Smarter

• How did special computer chips make AI smarter, and what problems still exist?
  
  CPUs are like doing homework one question at a time. GPUs are like having 1,000 friends help you all at once!

  1. What changed: Computers can now think about many things at the same time (like recognizing pictures, understanding words, and creating art)
  2. What's possible now: Computers became 10,000 times more energy-efficient in just 8 years!
  3. Limitations: We still need lots of electricity to run these smart computers, and we need to make sure they don't make mistakes that could hurt people

  The biggest challenge is making computers that use even less energy while being even smarter.
• Is it better to build AI chips that can do many things or just one thing really well?
  
  Imagine you have two toys:

  1. A Swiss Army knife that does many things okay (flexible)
  2. A super-duper scissors that only cuts paper but does it amazingly well (specialized)

  NVIDIA believes making flexible computer chips is better because:

  1. New ideas keep coming: Scientists always invent new ways for computers to think
  2. Nothing stays the same: Today's popular transformer AI (the brain behind ChatGPT) will change and improve
  3. More possibilities: Flexible chips let people experiment and create wonderful new things nobody has thought of yet

  Remember how highways created suburbs, gas stations, and fast food? Flexible technology creates whole new worlds of possibilities!
• How will computers keep getting more energy-efficient, and what cool things will that allow?
  
  Imagine if your night light got 10,000 times more efficient - it would use less energy than a firefly!

  1. What happened: Since 2016, AI computers got 10,000 times more energy-efficient while getting 6 times more powerful
  2. How it will continue: Engineers will keep designing chips that use less electricity while doing more work
  3. What will be possible: When computers get even more efficient, we'll see:
     • Helpful robots everywhere - lawn mowers, cars, and even humanoid helpers
     • Digital copies of human biology to help discover new medicines
     • Weather predictions that can see storms coming from miles away
     • Your own personal R2-D2 that follows you around and helps you

  Think of these super-efficient computers as "time machines" that let us solve problems in our lifetime that would have taken hundreds of years before!

Robots, AI, and How They'll Change Our Future World

• How will our lives change when robots do all moving jobs?
  
  When robots do all moving jobs:

  1. No more boring tasks: Robots will mow lawns, clean houses, and drive cars so humans don't have to.
  2. Your own robot friend: You might have something like R2-D2 from Star Wars that helps you every day.
  3. More free time: When robots handle physical work, you can focus on fun, creative things.

  To prepare, learn how to talk to AI helpers (like asking good questions) and think about what special human skills you want to develop!
• How do digital training worlds help make smarter robots?
  
  Digital training worlds help robots get smarter faster:

  1. Practice makes perfect: Instead of robots learning slowly in the real world, they can practice millions of times quickly in computer worlds.
  2. Learning physics: Omniverse creates realistic digital worlds with gravity and physics, while Cosmos helps AI understand basic things like "objects fall when pushed."
  3. No more oopsies: Robots can try difficult tasks in these digital worlds without breaking anything real!

  It's like how pilots train in flight simulators before flying real planes - robots get lots of practice before helping us in real life.
• What rules should we make for robots that will live with us?
  
  Important rules for robots that will share our homes:

  1. Safety first: Robots should have multiple backup systems to prevent accidents, like airplanes have extra safety features.
  2. Tell the truth: Robots shouldn't make up fake information (what grown-ups call "hallucinations").
  3. Be fair to everyone: Robots should treat all people equally and not have biases.
  4. No pretending: Robots shouldn't trick people by pretending to be specific humans.

  Just like we have driving rules to keep roads safe, we need robot rules to make sure our robot helpers are trustworthy friends!

How AI Works and Why We Need to Keep It Safe

• How can we fix the different problems with AI that might make it unsafe?
  

  AI can sometimes make mistakes, just like humans, but we need to be extra careful. Here's how we fix different AI problems:

  1. When AI makes stuff up - We give it true information from books and websites
  2. When AI breaks down - We build backup systems, like how airplanes have extra engines
  3. When AI might be hacked - We create security guards to protect it

  Think of it like having safety rules for a playground - we need rules to keep AI safe too!

• What can airplane safety teach us about making AI safer for everyone?
  

  Airplanes are super safe because they have lots of backup plans, and we can use the same ideas for AI:

  1. Multiple copies - Planes have three computers doing the same job, just in case one breaks
  2. Human oversight - Planes have two pilots watching over the computers
  3. Team safety - Air traffic controllers help pilots from the ground

  Like how many people work together to keep planes safe, we need teams of humans and computers working together to keep AI safe. It's not just one safety button, but many layers of protection!

• How do we make AI better and safer at the same time?
  

  Making AI better and safer together is like learning to ride a bike - you want to go fast but not crash!

  1. Test in pretend worlds first - Just like testing toys before giving them to kids
  2. Learn from mistakes - When AI makes an error, fix it before moving forward
  3. Follow core values - Jensen believes in building things that help people do amazing work, not replace them
  4. Start small - Try new AI ideas in safe places before using them everywhere

  The most important thing is remembering that AI should give people superpowers to do good things, not take over their jobs!

How Computers That Think Help People Do Amazing Things

• How will work change when AI does boring stuff for us?
  
  Imagine if your toys could clean up your room while you build cool LEGO creations! That's what AI will do for grown-ups:

  1. AI will do boring chores like sorting emails or writing simple reports
  2. People can spend more time thinking and creating instead of doing repetitive tasks
  3. We'll focus on what humans do best - having new ideas, solving tricky problems, and connecting with other people

  Just like having a helper who handles the boring parts so you can have more fun with the important stuff!
• How can AI helpers make us like superheroes?
  
  Think about how strong you feel when riding a bicycle compared to walking! AI will give our brains "wheels" to go further than ever before:

  1. AI acts like super-smart helpers that know lots of information instantly
  2. We can solve bigger problems because AI handles complicated parts for us
  3. Like having a team of expert friends who help us do things we couldn't do alone

  You're still you, but with amazing AI tools, you can learn faster, create more, and solve bigger problems - just like how Iron Man's suit makes him stronger than he could be by himself!
• How should schools teach kids to work with AI helpers?
  
  Schools need to teach kids how to be good "AI teammates" instead of just memorizing facts!

  1. Learn to ask good questions - Just like asking a friend for help, you need to know how to talk to AI clearly
  2. Focus on creative thinking - AI can find information, but humans need to decide what to do with it
  3. Practice working with AI tools - Kids should learn to use AI like they learn to use calculators
  4. Understand what AI is good at - Know when to use AI help and when to rely on human skills

  The most important school subject might become "How to be super-smart with your AI helper"!

How AI Will Change Our Future in Amazing Ways

• How will AI make science super amazing in the next 10 years?
  
  AI will be like a super-powered helper for scientists in the next 10 years!

  1. For biology: AI will help us understand how our bodies work much better. It will be like having a tiny map of how all the parts of your body talk to each other.
  2. For weather: AI will predict exactly what the weather will be like right above your house! Not just "it might rain" but exactly where and when.
  3. For medicine: AI will help doctors find new medicines much faster by understanding the "language" of proteins (tiny building blocks in your body).

  These things seem like magic now, but soon they'll be normal thanks to AI!
• What cool things could we do with computer copies of our bodies?
  
  Digital twins are like having a perfect copy of you inside a computer. Here's what we could do with them:

  1. Personalized medicine: Doctors could test medicines on your digital twin first to see exactly what works for YOU, not just what works for most people.
  2. Predict health problems: Your digital twin might show signs of getting sick before your real body does, so you could prevent illness before it happens!
  3. Learn about our bodies: Scientists could ask the digital twin questions like "What does this part do?" and it would explain how our complicated bodies work.

  It's like having a practice version of yourself that helps keep the real you healthy!