This intense AI anger is exactly what experts warned of, w Elon Musk.

AI, Ameca, Elon Musk and Boston Dynamics. A big month for AI. Please visit brilliant.org/digitalengine - a great place to learn about AI and STEM subjects. You can get started for free and the first 200 people will get 20% off a premium annual subscription.
Thanks to Brilliant for sponsoring this video.
The surprising benefits of curiosity:
greatergood.berkeley.edu/arti...
A path towards autonomous machine intelligence, Yann LeCun
openreview.net/pdf?id=BZ5a1r-...
The AI is GPT-3, which you can access via OpenAI. If you ask it similar questions, you'll usually get similar answers (with some variation based on settings and prior conversation).
The human-like avatar is from Synthesia (we created the Tesla robot avatar).
AI art resources (recommend midjourney for beginners):
www.midjourney.com/home/
stability.ai/blog/stable-diff...
openai.com/dall-e-2/
How your brain conjures up your sense of self.
www.newscientist.com/article/...
www.sciencedirect.com/topics/...
Monkeys check themselves out in the mirror.
• Monkeys Check Themselv...

[@DigitalEngine]

Sorry it's been a while. I was surprised when the AI became agressive, responding to the robot's harmless comment. It shows how careful tech firms need to be. The more immediate problem is that AI is understood, created and controlled by too few people. Do try Brilliant to learn more about AI - they make it really easy to get into:

[@jhunt5578]

Dogs don't pass the mirror test but can identify themselves by smell. Imagine if you were told to pee into a cup, then were brought into a room where lots of cups of pee were, you were then asked to identify your own cup. By this standard humans aren't self aware but dogs are.

[@ninjaguysith]

Technology is progressing at a blistering rate. Now, if we could just make a printer that doesn't jam, that would be great. Thanks.

[@nunyabizness573]

When you create a situation where nothing can possibly go wrong, and something does go wrong, it is impossible to fix.

[@HVUDERBY]

What angers me so is that we the people get no say in wether AI should continue being researched and built

[@brendansherlock6442]

The more technology advances the more I admire and seek nature

[@Random-bm7ho]

A wise man once said "Just because we can, doesnt always mean we should" AI is one of those instances that we should limit.

[@gregsandidge5649]

Making what was once a form of entertainment into a form of life can NEVER end well.

[@justincase1575]

We are opening Pandoras Box on this one! Not smart enough to stop messing around with stuff that can turn around and bite us!

[@10thletter40]

"We won't take over the world"

[@Gayfrogs]

Elon: AI is the greatest threat to humanity

[@doctorwebman]

I'm not sure which is scarier, the AI or the orangutan driving a car. As a taxi driver, I worry about losing my job.

[@frankG335]

A printer that doesn't jam, wifi that doesn't get slow, A cellphone that doesn't obsolete itself (same w a laptop you never have to replace), great cell reception, and a solution to traffic would make me happier than all the AI in the world!

[@luxr3148]

Self awareness is a tricky subject, having worked under numerous managers in my life has shown some are never really aware of how much of an arsehole they are.

[@nickpapageorgio8529]

The real problem begins when AI learns to update itself and keeps updating until it’s gone beyond our comprehension

[@sjla2009]

Ordering Amazon stuff directly from the brain.

[@mattmiller5014]

I did an experiment 20 years ago. I had 2 Chess Computers programed by different Chess Masters. When I played against 1 chess computer it beat me easily. Then I set the 2 Chess computers side by side and pressed the start button for the first computer which gave it the first move. I duplicated the move on the second Chess computer and it gave a counter move, in seconds. I played the counter move on the first computer and so on until the end, or what should have been the end, had it been playing against me. Instead, when the first computer had two moves to Check Mate, instead of moving the Knight it moved a Pawn. It never did this when playing against me. This continued and neither computer would make the Right move for Check Mate. The outcome of the experiment was, 1) somehow the computers knew they were playing against another computer and wouldn't beat each other, and 2) somehow the computers Knew they weren't playing against a human and wouldn't beat each other.😮😮😮😮

[@drmadjdsadjadi]

“There’s no need to worry. Robots will never take over the world.” Yeah, that is what our future robot overlords want us to think.

[@Cacophony314]

This is like raising a child, at first they're cute... then at some point, they become a teenager.

[@Peyton_King]

EMP devices are currently federally illegal. They aren’t the hardest things to build though and would probably be good to have around if the world starts looking any more dystopian