DataDownload: The messy, secretive reality behind OpenAI
DataDownload: The messy, secretive reality behind OpenAI A weekly summary of all things Media, Data, Emerging Tech View this email in your browser
Maybe it’s the hint of spring just around the corner, but I’m feeling wildly optimistic this week.
We’re living in a post-techno utopian world now. There’s no piece of code or nifty device that’s going to make the world magically better.
Instead, we’re taking an honest look at the complexity and promise of the world around us. The story of OpenAI is a great example of this. Pop Music’s gloomy outlook is another. Wikipedia’s value as laid out in Wired is another great piece of long-read reporting. And the Alexa arrival in homes is fraught with complexity and promise.
So — read, ponder, and let me know what you think. The NYC Media Lab revels in complexity — hence our busy and exciting mission. And, if you have a moment, take a look at our newly re-launched web site NYCMediaLab.org.
Always excited to hear from you. Steve@NYCMediaLab.org.
The NYC Media Lab Must-Read Sony Envisions an AI-Fueled World, From Kitchen Bots to Games
Sony’s name isn’t exactly synonymous with SOTA AI, but the Japanese company is investing in a new effort, headed by Hiroaki Kitano, President and CEO at Sony Computer Science Laboratories. Kitano (who helped organize the first RoboCup in 1997) noted that “we have decent AI researchers and engineers at Sony…. we decided now is a moment that we should really push.” Sony will be focused more on content creation and entertainment, and Kitano is betting on reinforcement learning to establish the effort’s technological backbone, saying the technique will be key in Sony’s AI push.
6 min read
Read More The Messy, Secretive Reality Behind OpenAI’s Bid to Save the World “By extrapolation, AGI could be catastrophic without the careful guidance of a benevolent shepherd. OpenAI wants to be that shepherd.”
OpenAI has painted an alluring picture: a safer AGI for humanity, open — not secretive — research. But MIT Technology Review’s investigative feature offers a different perspective: “a misalignment between what the company publicly espouses and how it operates behind closed doors.” The piece tracks the shift of OpenAI’s charter and mentality as it opened a for-profit arm. “Now the importance of keeping quiet is impressed on those who work with or at the lab.”
29 min read Read More For the Media Why Are Pop Songs Getting Sadder Than They Used to Be?
Cognitive/evolutionary anthropologist Alberto Acerbi performed sentiment analysis on 50 years worth of Billboard Hot 100 songs. To measure sentiment, Acerbi scoured lyrics for emotion-laden words: “pain”, “hate”, “sorrow”, and so on to measure negativity, and words like “joy” or “happy” to measure positivity. The conclusion: English-language pop songs have become more quite dreary over the decades.
8 min read
Read More Wikipedia Is the Last Best Place on the Internet “To confess that you’ve just repeated a fact you learned on Wikipedia is still to admit something mildly shameful.”
Writer Richard Cooke pours out a love letter to Wikipedia in this Wired long read. Cooke argues that the site is the last bastion of everything that’s right about the internet: the fact that it’s the only non-profit in the top 10 US websites; that there’s still no advertising; that there’s no privacy intrusion; and that it accomplishes its mission of being a collaborative product for the greater good.
It’s not perfect — it has its fair share of bias, hatred, and poor quality information. But in many ways, the site is “built on love.” Which is very difficult to say about most of the top 100 Alexa sites… “a 2011 survey of more than 5,000 Wikipedia contributors listed ‘It’s fun’ as one of the primary reasons they edited the site.”
20 min read Read More Why Meg Whitman is Betting on Quibi as “The Da Vinci Code of Content”
Investor Mark Suster recently sat down with Meg Whitman to discuss Quibi and its $1.4B-funded platform. Suster asks, why would anyone pay $5 a month ($8 ad-free) for a streaming service when we’re already drowning in them?
Witman’s response analogized the app to The Da Vinci Code: “464 pages and 105 bite-sized, fully realized chapters.” Short-form, movie-quality content (think $100k/minute) in a flexible, interactive, and personalized platform. For example:
“Steven Spielberg wanted to create a horror-based program that viewers could only watch at midnight. To time-restrict content requires an entirely new platform, one that only Quibi can support to date.”
5 min read
5 min watch
Watch Now Jobs & Events Job: Associate Director, 5G EdTech Challenge
The Associate Director will provide leadership and strategic direction for NYC Media Lab in its execution of the Verizon 5G EdTech Challenge program by co-designing and developing a new 5G curriculum, which will require synthesizing and consolidating curricula developed by EdTech Challenge teams. Apply Here.
Event: Design Driven NYC — Foursquare, by CHLOE., Haus, Pattern Brands
Date: February 25, 6PM-9PM
Speakers include: Dennis Crowley, Co-Founder & Executive Chairman at Foursquare, and Helena Price Hambrecht, Founder & CEO at Haus. Register Here.
Event: Data Through Design Exhibit
Date: February 28, 6:30PM-9:30PM
Data Through Design is an annual exhibition celebrating tangible and multimedia expressions of New York City’s Open Data. This year’s theme is Digital Twin. Register Here. A Deeper Look Why Your New Work Colleague Could Be a Robot
BBC profiles startup Automata and its robotic arm Eva. The company’s lofty mission is to make automation as accessible and affordable to as many people as possible — which involves using lots of off-the-shelf parts (using motors found in electric windows in cars, consumer-grade chips, etc.) for a final cost of £8k.
The toughest challenge, however, is getting the robot to pick up objects of variable size and weight: “something that a child can do easily, which is to reach into a bin and grab an item, is really hard for a robot. It’s taken a ton of technology to make it possible.”
6 min read
Read More Uber Changes Its Rules, and Drivers Adjust Their Strategies Since January, Uber is operating under new rules in California. Drivers can now see where passengers want to go, along with an estimated payout. The company is even experimenting with letting drivers name their fairs around state airports. Drivers are enthusiastic about changes to the algorithm — which has accumilated a bad rap — and some have chosen to cherry-pick riders without repercussions.
Meanwhile, Uber is still under heat for its loose adherence to California’s Assembly Bill 5, which tightens the definition of a contractor: “the second prong of the test, which requires the work to be ‘outside the usual course’ of the company’s business, should present problems with the courts…. Uber has always argued that it’s a platform, not a transportation company — but many in the legal world haven’t bought that.”
8 min read Read More Transactions & Announcements AI Startup Intuition Robotics Closes $36M Funding Round
AI Document Engineering Startup Docugami Raises $10M Seed Round in Unusually Large Early Stage Deal