The following article and talk are by Jessica Taylor and they were originally posted on MIRI. At the EA Global 2016 conference, I gave a talk on "Using Machine Learning to Address AI Risk": It is plausible that future artificial general intelligence systems will share many qualities in common with present-day machine learning systems. If so, how could we ensure that these systems robustly act as intended? We discuss the technical agenda for a new project at MIRI focused on this question. The talk serves as a quick survey (for a general audience) of the kinds of technical problems we're working on under the "Alignment for Advanced ML Systems" research agenda. Included below is a version of the talk in blog post form.1 This talk is about a new research agenda aimed at using machine learning to make AI systems safe even at very high capability levels.

A few months ago I wrote a tutorial on how to classify images using Convolutional Neural Networks (specifically, VGG16) pre-trained on the ImageNet dataset with Python and the Keras deep learning library. The pre-trained networks inside of Keras are capable of recognizing 1,000 different object categories, similar to objects we encounter in our day-to-day lives with high accuracy. Back then, the pre-trained ImageNet models were separate from the core Keras library, requiring us to clone a free-standing GitHub repo and then manually copy the code into our projects. This solution worked well enough; however, since my original blog post was published, the pre-trained networks (VGG16, VGG19, ResNet50, Inception V3, and Xception) have been fully integrated into the Keras core (no need to clone down a separate repo anymore) -- these implementations can be found inside the applications sub-module. Because of this, I've decided to create a new, updated tutorial that demonstrates how to utilize these state-of-the-art networks in your own classification projects.

Machine learning couldn't be hotter. A type of artificial intelligence that enables computers to learn to perform tasks and make predictions without explicit programming, machine learning has caught fire among the hip tech set, but remains a somewhat futuristic concept for most enterprises. But thanks to technological advances and emerging frameworks, machine learning may soon hit the mainstream. Consulting firm Deloitte expects to see a big increase in the use and adoption of machine learning in the coming year. This is in large part because the technology is becoming much more pervasive.

Approach-avoidance may be the best way to characterize Americans' attitudes towards a driverless future. Three-quarters (78%) of U.S. drivers reported feeling afraid to ride in a fully self-driving vehicle, yet most of them --59% -- said they want autonomous vehicle technology in their next vehicle. So while American drivers seem ready embrace autonomous technology, they are not yet ready to give up full control. Those are the main findings a new survey released earlier this month by AAA. "A great race towards autonomy is underway and companies are vying to introduce the first driverless cars to our roadways," Greg Brannon, AAA's director of automotive engineering and industry relations, said in a statement.

What kind of existential problems does AI bring about? The medium-term challenge of AI is not killer robots, it's job replacement. This dynamic is already underway and the literature suggests it's a more powerful driver of job loss than trade, though trade receives much more attention. True AI has not arrived, and automation is not AI, but robots and human-written code are a reasonable preview of what employment challenges genuine AI will bring. Computers already manage warehouses, can drive reasonably well, and are making meaningful progress into areas like basic lawyering and radiology that we long considered to be immune to change.

HBO's Westworld features a common plot device--synthetic hosts rising up against their callous human creators. But is it more than just a plot twist? After all, smart people like Bill Gates and Steven Hawking have warned that artificial intelligence may be on a dangerous path and could threaten the survival of the human race. They're not the only ones worried. The Committee on Legal Affairs of the European Parliament recently issued a report calling on the EU to require intelligent robots to be registered, in part so their ethical character can be assessed.

In the past years, a collection of hardware, software and online service have managed to bring changes and reforms to classrooms and teaching methods. But the true disruption of education is yet to arrive. Artificial Intelligence has proven its role as a game changing factor in an increasing number of fields, causing transformations unimaginable in the past. It's now showing glimmers of how it might forever change the learning process, one of the oldest skills that mankind has mastered. Gary Vaynerchuk was so impressed with TNW Conference 2016 he paused mid-talk to applaud us.

Machine learning couldn't be hotter. A type of artificial intelligence that enables computers to learn to perform tasks and make predictions without explicit programming, machine learning has caught fire among the hip tech set, but remains a somewhat futuristic concept for most enterprises. But thanks to technological advances and emerging frameworks, machine learning may soon hit the mainstream. Consulting firm Deloitte expects to see a big increase in the use and adoption of machine learning in the coming year. This is in large part because the technology is becoming much more pervasive.

IBM announced an important milestone in conversational speech recognition last year. The company managed to develop a system that achieves a 6.9 percent word error rate. Despite the success, IBM continued to work hard on its speech recognition technology and has recently achieved a new industry record of 5.5 percent. In an official blog post, the company said that the word error rate was measured with the help of recorded conversations between people discussing usual everyday topics like buying a car. These recordings, which are known as the "SWITCHBOARD" corpus, have been used in the industry to benchmark speech recognition systems for more than 20 years.

Among all of the self-driving startups working toward Level 4 autonomy (a self-driving system that doesn't require human intervention in most scenarios), Mountain View, Calif.-based Drive.ai's Drive sees deep learning as the only viable way to make a truly useful autonomous car in the near term, says Sameep Tandon, cofounder and CEO. "If you look at the long-term possibilities of these algorithms and how people are going to build [self-driving cars] in the future, having a learning system just makes the most sense. There's so much complication in driving, there are so many things that are nuanced and hard, that if you have to do this in ways that aren't learned, then you're never going to get these cars out there." It's only been about a year since Drive went public, but already, the company has a fleet of four vehicles navigating (mostly) autonomously around the San Francisco Bay Area--even in situations (such as darkness, rain, or hail) that are notoriously difficult for self-driving cars. Last month, we went out to California to take a ride in one of Drive's cars, and to find out how it's using deep learning to master autonomous driving.