Some really spooky AI [MIT Media Lab]

“I woke up in the hospital the next day. My one friend told me that something was wrong with me. I began to get the feeling I was being watched by something I couldn’t see”. Spooky, no? That’s the first line of a horror story written by Shelley, an AI trained by MIT researchers to write horror stories. Named after Frankenstein author Mary Shelley, it was trained by processing thousands of subreddits – threads on Reddit – of original horror stories. While Shelley cannot invent stories from scratch, it’s clearly adept at turning human inspiration into machine prose. Fun tip: anyone can contribute to Shelley and tweet responses that the AI might pick up!

Tech Team


Will robots have more rights that human beings? [Inc]

Are robots human equals? That’s what we’re asking ourselves after Saudi Arabia granted a female humanoid robot named Sophia citizenship. For some, this announcement proved the country is modernising. But many pointed out the robot had more rights than migrant workers or Saudi women, as it wore no headscarf and abaya and didn’t have a male companion. Beyond the specific gender issues of Saudi Arabia, this raises ethical problems particularly regarding the place we want to give to robot in our societies. While it may seem like a PR stunt today, robot rights are quickly becoming an issue that must be tackled.

The sharing economy enters the closet [The Guardian]

Forget the walk-in wardrobe as the lifestyle must-have for the fashion conscious. The sharing economy that has revolutionised our approach to transport, accommodation and many more aspects of daily life is now shaking up attitudes to our wardrobes. As with cars and holiday homes, ownership of clothes is no longer considered essential. Thanks to the internet, New York-based Rent the Runway and UK business Girl Meets Dress are filling a niche generated by living space constraints and the appetite for instant fashion gratification.

Does the US government have the smarts to regulate AI? [The Next Web]

AI is a topic that has seen much speculation, especially in the past year. Remember Elon Musk’s “it might cause WWIII” tweet? The Next Web’s Tristan Greene argues that such scaremongering and hastily drafted new legislation might actually cause serious damage. How? Strict new regulations rushed out by panicked politicians without experience or education on AI could lead to laws that stifle its development. Instead, he suggests, let the experts develop AI within the framework of law and encourage more robust innovation supported by ethical guidelines.

China revving up its car industry’s computing power [Financial Times]

An interesting shift and a new race are emerging among China’s tech giants about the development of car operating systems or what we’d call smartphones on wheels. The race to build intelligent and connected vehicles is on, and China’s leading internet groups – Baidu, Alibaba and Tencent – are trying to develop driving-related software. Why? More than 60% of Chinese consumers are willing to switch car brands for better in-car connectivity. The day we will be choosing cars depending on software rather than hardware might not be that far off. Horsepower will remain key but… for computing!

In case you haven’t had enough:

Nvidia CEO: “No more than 4 years to have fully autonomous cars on the road” [The Next Web]

Watch Hackers Sabotage an Industrial Robot Arm [Wired]

Artificial intelligence: Silicon Valley’s new deity [Financial Times]

Who Will Rein In Facebook? Challengers Are Lining Up [Wall Street Journal]

How Germany’s Otto uses artificial intelligence [The Economist]

Best-Ever Algorithm Found for Huge Streams of Data [Wired]


#TechAways is brought to you by Cambre’s Technology Practice led by Victoria Main and featuring François BarryZachery BishopSvenja MaiAnne-Claude MartinSimos Piperidisand Teodora Raychinova.

Questions, comments or ideas to