The Cambrian explosion in robotics; wind power’s bright future; the paradoxical science of gender fluidity; climate change in Syria; AI design ethics; Silicon Valley’s nouveau riche. Shorter, as promised!
Forwarded by a friend? Subscribe
Dept of near future
👻🐝🐒 Are we on the verge of a Cambrian explosion in robotics? By DARPA’s former robotics honcho, Gill Pratt, who has recently been hired by Toyota. Detailed, worth reading
🍔 DRM for food? Some restaurants really dislike neutral-delivery platforms and want a sort of DRM for meals: you can buy it but you can’t sell it, give it away or ask someone else to pick it up for you.
🍃 Wind power can have a bright future and avoid turbulence, argues Ramez Naam. Great primer
🎒 “Udacity is my response to the development of AI. The mission I have to educate everybody is really an attempt to delay what AI will eventually do to us, because I honestly believe people should have a chance.” Great profile of Sebastian Thrun in The Economist.
👫 Emerging science supports the basis of gender fluidity, but our neurobiology is strongly weighted towards fast binary classifications. Excellent read
😠 The roots of Syria’s hideous conflict can be traced to anthropogenic climate change. Cartoon exposition
😮 “Humanities dumbest experiment” Elon Musk on our love of fossil fuels Punchy
😎 Startup L Jackson is the nom-de-plume for an insightful, acerbic commentator on the innovation factory that is Silicon Valley. Here is a rare Q&A with him. Funny
Dept of over optimisation, data & learning
At a SOMA Salon last month, I presented on the ethics of design AI-based systems. Traditional design thinking is clearly an important part of designing products that use AI. But there are some special considerations that apply to AI-based systems: learning from the world as is and over-optmisation.
Two recent posts demonstrate this.
This research group attempted to predict runway model success by looking at wide slews of data about models. It’s a reasonable example of the kinds of new approaches internet-connected data can afford the prediction problem. The researchers joined together disparate data: physical attributes, contractual elements, Instagram behaviour and even sentiment on social. Then applied this to what has traditional been an ambiguous problem of cultural filtering.
The paper, which is limited in its scope and conclusions, highlights many of the uncomfortable facts of machine learning: (1) we can only represent and reinforce the world as it was (2) the algorithm & feature choice determines the nature of the story we can tell (3) AIs are systems optimised for ruthless optimisation, so any biases in (1) and (2) will get magnified.
Undoubtedly the demand for computer-mediated filtering of social or cultural processes will increase. A more comical example is how these “horny nerds” used deep learning to identify attractive women on Tinder. Here machine learning or AI was used to reduce friction in a process, in this case mate or partner selection.
Providing both a challenge and an opportunity is that much of the data we’ll need to explore the world is no longer publicly accessible. Instead it is held in the data centres of some of the world’s largest technology firms.
The good news: these firms are steeped in the language of experimentation and so do at least exploit these data at an experimental scale that would make most in the academy jealous, as Michael Schrage argues.
The bad news (as I argue) is that these commercial experiments are optimisations around very particular local maxima: that of generating whatever outcome a particular product needs to achieve (e.g more clicks on adverts, longer user-linger time to create more opportunities to show adverts, higher referral loops to drive more users to ultimately increase advertising inventory, and so on.)
Perhaps one day we’ll see large firms offer access to these data and their experimental infrastructure to improve academia’s ability to explore and test wide-ranging rather than commercial hypotheses.
Why deep learning will lead to troublesome questions in art.
Short morsels for dinner parties
💰 Nick Bilton’s Vanity Fair article asks if we are in a bubble. Interesting expose of the weirder excesses of current Silicon Valley (like the secret group of Facebook execs called ‘The Nouveau Riche’).
The quest for scale is crushing local & community news. There is demand for local and community news but the economic of traditional media in the digital world don’t support it. Cracking community information & discussion becomes a big opportunity, especially when we combine it with the new localism of pop-ups, neo-artisans and craft.
A record year for batteries.. Industrial energy storage/battery projects are up 9x YonY. Even US residential battery demand grew 61% QonQ.
Eye-contact is an important intention transmission modality amongst human drivers. Self-driving cars can’t read your glances yet. Interesting challenge.
More details on Toyota’s major investment in uncrashable robo-cars.
The first autonomous killer robots in operation are hunting down destructive star fish in coral reefs. (Video.)
Xerox scanners randomly alter numbers in documents they scan. This researcher thinks is due to a bug in the lower-levels of the image encoding.
Alaska’s glaciers are losing 75 gigatons of ice every year.
🐙 The remarkable octopus: “an intelligent animal with entwining arms so filled with neurons that each of them possesses a separate personality”. I stopped eating octopus earlier this year as I became more aware of the science demonstrated its much higher brain functions including, potentially, 'theory of mind’.
There are three trillion trees on the planet. Humanity has axed a further three trillion.
😕 The British Pyschological Society reports how a commercial tDCS headset reduces rather than improves working memory. tDCS headsets zap the pre-frontal cortex with electricity and are popular amongst amateur brain hackers, e.g. on Reddit. Neuromodulation is a scientifically verifiable thing, see Amol Sarva’s company, Halo, for example.
䷕ Lovely, practical guide to what type of machine learning can answer a given business question. Helpful for product managers or execs trying to get something useful out of their data scientists.
What you wrote
AEV Subscriber John Hagel of Deloitte, argues that new economics of distribution will support the growth of independent fashion designers as they get the ability to access and serve huge markets without retailers.
Thanks for filling out the feedback last week. As always ping me for chat / discussion.