This is a talk by Ilya Sutskever for course 6.S099: Artificial General Intelligence. He is the Co-Founder of OpenAI. This class is free and open to everyone. Our goal is to take an engineering approach to exploring possible paths toward building human-level intelligence for a better world.

DeepMind created an AI which spontaneously developed the machine learning equivalent of gut-based navigation. The UK-based Google sister-company seems to specialize in creating machine learning experiments designed to determine if AI can inform the field of neurology, and vice versa. DeepMind recently published a paper demonstrating a neural network that, upon trying to solve a navigational problem, developed a method of spatial awareness that imitates the creation of "Grid Cells" in mammals. Grid Cells, which were discovered in 2005, are a little-understood phenomena that occur within mammal brains to aid with navigation. Basically, our ability to generally understand where we are based on how far we've traveled and in what direction, is governed by these specialty cells that form in hexagon-shaped patterns that the brain sort of overlays into space, causing neurons to fire when we move through it.

Google taught its DeepMind AI to remember things like a human would. Most AIs can specialize in one area, like defeating the world's best Go players; but DeepMind was programmed to apply previous knowledge and skills to learning new tasks, drawing from a neural network of programmed skills and "memories". Now, DeepMind is teaching itself how to organize its own "brain" network. And Google researchers were shocked when, without any input from them, the AI chose to make part of its brain look nearly identical to humans. Google's DeepMind team, in collaboration with University College London (UCL) researchers, stuck the AI in a virtual reality maze to teach it spatial awareness and memorization of patterns, publishing their findings in Nature.

LONDON, ENGLAND - DECEMBER 05: Co-founder of Google DeepMind Mustafa Suleyman attends a Q&A during day 1 of TechCrunch Disrupt London at the Copper Box on December 5, 2016 in London, England. Google has found another use for DeepMind's machine learning software after buying the London artificial intelligence lab for a reported £400 million in 2014. Later this year, the search giant will roll out two new DeepMind-built Android features that are designed to improve battery life and optimise screen brightness levels. The features will be available to people with devices running the Android P operating system. The features -- announced during the Google I/O developer conference -- were built by a unit called "DeepMind for Google," which focuses on applying DeepMind's technology to Google products.

Google has found another use for DeepMind's machine learning software after buying the London artificial intelligence lab for a reported £400 million in 2014. Later this year, the search giant will roll out two new DeepMind-built Android features that are designed to improve battery life and optimise screen brightness levels. The features will be available to people with devices running the Android P operating system. The features -- announced during the Google I/O developer conference -- were built by a unit called "DeepMind for Google," which focuses on applying DeepMind's technology to Google products. The same unit has also helped Google to reduce energy use in its data centres, optimise recommendations in Google Play, and improve the speech for Google Assistant users and Google Cloud Platform users.

Artificial intelligence is winning the rat race. Google-owned DeepMind has built an artificial intelligence that is better at navigating a maze than humans. After it was trained with data on how rodents search for food, it mimicked the processes that allow mammals to get between destinations in the most efficient way. Humans and other mammals have neurons called "grid cells" that help us find our way as we navigate our surroundings.

Recorded May 2nd, 2018 at ICLR2018 Koray Kavukcuoglu is the Director of Research at DeepMind, where previously he was a research scientist and led the deep learning team. Before joining DeepMind, he was a research staff member at NEC Labs America in the machine learning department.

The following is adapted from State of Green Business 2018, published by GreenBiz in partnership with Trucost. There is no shortage of smart people willing to offer their sometimes dire, sometimes optimistic opinions about how humankind's future will be reshaped by computers and software using some sort of artificial intelligence (AI). If there's one thing upon which the naysayers and yeasayers agree, it's that AI is already more real than many people realize. A whopping 70 percent of the companies surveyed last year by Forrester Research plan to use some form of AI by the end of this year. It's tough to think of a tech giant that isn't making AI research a priority: Alphabet (through DeepMind and Google), Amazon, Apple, Facebook, IBM and Microsoft are throwing literally millions of dollars at this opportunity.

The tech giants already know this and are investing in democratizing AI to make tools and services more widely available, but the user experience (UX) of machine learning is still overlooked. Companies can make massive improvements to machine learning-based applications even without access to the same levels of data or talent as the biggest players -- compensating for a lack of data by building a great UI (more on this later). When we focus on AI as a tool and recognize how crucial usability is to widespread adoption, we can see that there are opportunities to enhance existing AI in ways that have nothing to do with progress toward human-level machine intelligence or artificial general intelligence. While flashy projects like DeepMind and Google Brain are more likely to make headlines than Google's more mundane implementations of AI, such as search, the latter is a vastly more profitable business. According to a recent MarketWatch article, Google has "made a massive multibillion-dollar bet on AI and machine learning," a bet I believe is nicely hedged on the question of whether there'll be another "AI winter," a period of reduced interest in AI.

What appears to be a single task actually depends on multiple cognitive abilities. We face a similar problem in AI research, where the complexity of a task can often make it difficult to tease apart the individual skills required for an agent to be successful. But understanding an agent's specific cognitive skill set may prove useful for improving its overall performance. To address this problem in humans, psychologists have spent the last 150 years designing rigorously controlled experiments aimed at isolating one specific cognitive faculty at a time. For example, they might analyse the supermarket scenario using two separate tests - a "visual search" test that requires the subject to locate a specific shape in a pattern could be used to probe attention, while they might ask a person to recall items from a studied list to test their memory.