What would you program your automaton to do? Will we even decide or will the machines decide for us? and what will we do then? Lot’s of ethics questions in today’s briefing but also a guide which shows how you can use Deep Learning to make film stars, dignitaries or even Mona Lisa to give you a wink.
Among other things, Musk is currently fueling the open-source, non-profit org OpenAI in its efforts to make automatons capable of teaching themselves to care for us through simple, high-tech trial and error.
Janet Burns, Forbes contributor explains “robotic helpers from industrial-supplier Fetch Robotics as hardware, OpenAI’s software developers have been working to create artificial intelligence capable of advanced machine learning through ‘deep reinforcement’ by observation and exploration, allowing such robots to develop and expand their own skill sets.”
As the MIT Technology Review reports, the mobile bots utilize such tools as a 2-D laser scanner, 3-D depth sensors, and a versatile grabbing arm to come to grips with the world around them and build up abilities in their neural networks as they figure out how to best perform a task.
Stephen Deangelis, of Enterra Solutions nicely summaries what a lot of people are hoping in this Wired article.
“Will it create new jobs? Certainly. In fact, I suspect that entire new business sectors are going to be developed as a result of AI. The question will be whether enough new jobs can be created to replace those that have been taken over by smart machines. Will artificial general intelligence be developed? Maybe. And if AGI is developed, that’s where caution needs to be taken. Is humankind in danger? Not at the moment. Narrow uses of AI are going to help humans do a lot of amazing things in the years ahead”
Juliette Powell’s interview of Andrew Ng, chief scientist of Baidu Research, takes us on a journey from AI being a larger socially valuable endeavor all the way to seeing it as the first step towards universal basic income, which would provide financial support for people who committed themselves to being lifelong learners. Maybe one for chats over dinner tonight? or maybe not…
Yes you read that correctly. This tool takes images and searches for eye features, and then, well makes them move. Perfect for when your day can be lightened by Ryan Gosling giving you “the eye” or for our Harry Potter reading fans, when you want to look feel like you’re in Dumbledore’s office.
From the makers of ‘Joy‘, the mental health chatbot we linked to in the Briefing on Tuesday, a pretty comprehensive guide to what you should and shouldn’t do while making a chatbot. From the use of emojis to come across as slightly human, to not being afraid of leading the user away from the chat interface (in this case to an online doctor consultation). It’s worth a read.
Article to share with your less Data Savvy friends
Charles Arthur, author or Digital Wars writes a really tweetable, understandable and comprehensive article to share with your friends who aren’t quite sure yet that Machine Learning is coming and it’s coming fast!
“If you think it’s not worth your business bothering with machine learning, consider what rivals might be doing.”
Neural Net turns NYC into moving painting: Imagine the world where you can choose how to see it through AR. Danil Krivoruchko’s new Machine Learning project has alloowed this video of New York to be changed into the style of a painting. How would you like to see the world? Answers on a postcard please.
The effectiveness of online marketing is all driven by serving the right ad, to the right person, at the right time to get them to interact. Google is pushing the boundaries all the time to make this more effective. They recently launched a new version of Google Ad Words with ML driven real time bidding on ad slots for marketing agencies and, of course, brands directly.
We’ll be testing it to reach you all – let us know 🙂
Ever wondered what a Facebook Data Scientist looks like? They are mythical beasts indeed. Luckily, while we were out playing Pokemon Go, we managed to capture one: Ahmed Medhat. He’ll be joining us, along with his friend Nadbor Drozd and real Data Science interview whizz at our next event on August 10th.
We are almost fully booked so make sure you sign up today