🔮 Can you trust AI? Zuckerberg's genius; future of economics; growth mindset; for & against basic income; dog IQ; emoji & deodorant++ #65
|Azeem Azhar||Jun 5, 2016|
Machine intelligence and consciousness special. Rethinking economics; the importance of breadth & depth; how will talking to our tech, change how we talk to people; basic income pros and cons; Man vs machine in the kitchen, on the road and on the psychiatrist’s couch; can neuroscience understand Donkey Kong; dog IQ; when deodorant works; and the lasting effects of poverty.
A rich seam this week, so keep going to the end.
SUPPORT EV: Donate via PayPal. (You choose the level.)
Dept of the near future
🎓 The importance of the T-shaped student, with breadth, depth & the growth mindset, in the future economy. INSIGHTFUL
🌟 When products talk: “We’ll likely learn to treat bots more like people. But, in the process, we may end up treating people more like bots”, Tyler Cowen in the New Yorker. THOUGHT-PROVOKING
💡 Why a basic income is necessary for the US, a small-government argument by Charles Murray. (The FT says it probably won’t work; John Kay | Alan Beattie. “Basically flawed” argues The Economist.) GOOD READS
😠 Why the world is drawing battle lines against American tech giants. Introducing the frightful five. (See also Hossein Derakshan: “Facebook’s algorithms control us, we need to resist.”)
👍🏽 “Do Believe the Hype (Benefits Coming Later)” Albert Wenger neatly sums up Carlota Perez’s framework on the phases of technology. SHORT, ILLUMINATING
Dept of pay-it-forward
😀 Exponential View is closing in on 10,000 readers. To say thank you to all of you, 25 readers can get a free copy of Sam Arbesman’s new book, Overcomplicated. Sam is Scientist-in-residence at Lux Capital.
Overcomplicated is about the forces that make technology more complex and more incomprehensible over time, despite our desire for this to be otherwise. How can we deal with these complex systems that surround us? Sign-up here.
Dept of machine intelligence
Can AIs solve creative or ethical problems as well as humans? These two areas seem to be bastions of comfort for Meat Intelligence, like us.
🍲 Chef, Yottam Ottolenghi is pitted against IBM Watson in this delightful essay in The Guardian. Yottam’s dish is better, no doubt.
Google announces project Magenta designed to investigate machine-generated music and art. They released their first 90-second AI-composed piano melody. Listen to it, and think of the Wright Flyer struggling across Kill Devil Hills. In any case, so much of our modern cultural output is systematised through the 🎶 trap-and-hook formula that dominates pop music.
IEEE Spectrum digs deeper into how to design ethical systems for self-driving cars. Should they be allowed to flout the laws? Probably.
What about helping us with therapy? A computer vision system is 3 times better at identifying real pain in a person’s face than a trained human. And we know that computer systems keep calm under pressure. Should we have them make snap triage decisions rather than us?
“Therapy chatbots have the potential to fall into a category that I call ‘means well’ technology” argues Natalie Kane. Who is to blame when they go wrong?
Simple technical intro: the difference between deep learning and machine learning.
Dept of consciousness
Two neuroscientists applied modern neuroscience techniques to a very basic silicon chip, the 6502 that powered the Apple II and a clutch of other personal computers through the mid 1980s. The 6502 is a tiny chip, with only 3510 transistors compared to 86 bn+ more complex neurons in the human brain.
🔬 The results were salutary. (HIGHLY ACCESSIBLE) They couldn’t make sense of the 6502 or how the software it was running, including games like Donkey Kong, interacted with the system. “If it doesn’t work on the chip, how can we expect it to work on the brain?”
Accessible review of a new study on how the brain creates consciousness “has identified a clear metabolic threshold for the emergence of consciousness”. It is 47% of normal metabolic activity, apparently.
Marcus Du Sautoy: “Computers could develop consciousness and may need ‘human’ rights.”
🐶 Of course, the same thinking could apply to animals, especially as we start to develop formal understanding of animal generalised intelligent, e.g. the canine equivalent of IQ. (Good read for dog lovers.)
Exponential Dinner #7 - Disrupting disruption
Exponential Dinners are private, participatory intimate dinners where EV readers can sit in conversation with a world expert & other readers on a relevant topic.
Our next dinner is Disrupting Disruption. Our interlocutor is EV subscriber Joshua Gans, Skoll Chair in Innovation & Entrepreneurship at the Rotman School. We will discuss innovation, disruption and scientific progress with the usual fab crowd.
Dinner is at 7pm on July 4th in London; the cost will be £75.
Short morsels to appear smart at dinner parties
🚀 How Zuck led Facebook’s war to crush Google. EXCELLENT READ (Vanity Fair, so generalist, not tech nerd). Read it alongside this short HBR piece on “What keeps companies growing”: insurgency, owner’s mindset and a front-line obsession. (Also see the Hossein Derakshan piece at the top of EV.)
Deodorant works. Helps low masculinity men raise their game.
The top 25 books in the social sciences. I’ve read five and touched on three. How about you?
“One 71-year-old wheelchair-bound patient was walking again” Injecting stroke-patient brains with stem cells.
Epigenetic effects of poverty: altering your DNA and increasing likelihood of depression.
And one to make you laugh for Sunday: The risks of strong AI.
What you are up to
EV reader, Prof Kevin Werbach, on conversation, the fifth-interface.
EV reader, Rodolfo Rosini, on WeaveOS, an AI-first operating system.
Jonathan Libov, EV subscriber, on design, art and culture at zero marginal cost.
Enjoy the sun, it is bbq weather in the UK.
Please continue with your amazing recommendations & email referrals!
One final ask - if you are using Gmail, please drag Exponential View from the ‘promotions’ tab into your inbox! :)