Microsoft Research Blog

The Microsoft Research blog provides in-depth views and perspectives from our researchers, scientists and engineers, plus information about noteworthy events and conferences, scholarships, and fellowships designed for academic and scientific communities.

Exclusive: Experts and audience agree AI is an opportunity, not a threat

December 23, 2015 | Posted by Microsoft Research Blog

By Christine Clifton-Thornton, Research News

Swedish Nobel organizers whose annual prizes recognize the pinnacle of human achievement turned their attention this month to human and artificial intelligence. The Nobel Week Dialogue convened leading academics, technology leaders, Nobel laureates and other luminaries in Gothenburg, Sweden to consider “The Future of Intelligence.” Capping a year of rising concern in some quarters of the dangers of AI, the event provided lively and thoughtful discussions on what remains a controversial topic.

“We design the systems we put in the world, and there are ethical issues that need to be understood and addressed,” said Harvard AI professor Barbara Grosz in a panel addressing the future development of AI. “We have a choice as we think about those issues, and many issues can be addressed by designing systems better.” And students should consider potential negative impacts of AI early. Grosz teaches a class on intelligent systems design and ethical challenges that engages students through the tool of popular media, examining movies such as Ex Machina under the light of ethics.

“As we design these kinds of systems, we have to be very careful to take into account their impact on human beings,” agreed Harry Shum, Executive Vice President of Microsoft’s Technology and Research group. “We have more and more data and breakthroughs in algorithms that have allowed us to make great progress in speech, language and understanding, and I personally believe Microsoft must invest more in these areas.”

“Until now, the tasks and goals we want AI systems such as Watson to solve have been very specific,” said Shum. “But we continue to struggle with our inability to model uncertainty. We really don’t know which is more difficult: Landing a drone on an aircraft carrier or designing a little robot to make eggs for your breakfast in your kitchen. I think it’s the latter, because we don’t know how to model the differences among houses. But we’ll understand more and more, largely from the data we collect.”

Perhaps nothing better illustrates the rapid change in technology and intelligence than the increasing amount of data generated each day. About 90 percent of the world’s data has been created in the last two years alone, with 2.5 exabytes of new data being created every day, according to Guru Banavar, Vice President of Cognitive Computing at IBM and leader of the research team creating the next generation of Watson.

And what should we do with this data? How about developing a learning, reasoning machine that would partner with people? Banavar suggests such a machine would be a powerful tool to augment human intelligence—not replace it.

Anyone concerned about the imminent rise of the machines need only view some recent contests of machines in action. “At the DARPA competition this year, the robots spent most of their time just trying to get up,” said UC Berkeley professor Stuart Russell.

Future predictions

When asked to predict the ten-year future of AI, panelists envisioned a broad range of useful systems. “We should provide advice and expertise to enhance people’s ability to participate in government and bring what people at large think to the government,” said Grosz.

“One of the things machines can do really well is collect and synthesize large amounts of information,” said Russell. “Without machines, we don’t have the ability to put together information—artifacts, texts, all kinds of historical sources—showing how our cultures have grown and interweaved. We should use the information to build a consistent consensus picture about the functioning of life, and do the same thing with history.”

“There should be an all-knowing agent for you: an alter ego, your second self,” predicted Shum. “But it does require the user to cooperate with the AI system: we can only build that if you are willing to share. I am a big believer in having the data. I think the alter ego will be the most exciting AI system in the future.”

But we don’t have to wait for the future to see the positive impacts of AI. “With recent breakthroughs in algorithms and science, AI is already making a difference in people’s lives,” said Grosz. Current uses of AI are legion. “It’s being used to protect wildlife in the forest by understanding the randomness of where the poachers are going to be.” And opportunities in the healthcare field abound. “In the US alone, every year 400,000 people die from medical errors,” said Shum. Medical applications, some already in development, can prevent such errors.

Nobel a trending topic

The Nobel mission of bridging science and society proved to be a hit with its core audience. The digital dialogue was initiated before the event began, when thousands of Nobel social media followers responded to a question on the Future of Intelligence that helped inform the discussion in Gothenburg. “It’s the first time the Nobel Week Dialogue has made it to number one,” said Magnus Gylje, Editor-in-Chief of, noting that #NobelDialogue was the most-discussed topic in Sweden throughout the day on Twitter and trended internationally.

During the Dialogue, the audience responded to live polling questions such as, “Is the development of AI primarily a threat or an opportunity to our society?” The verdict: More than 86 percent believe AI is an opportunity.

With so many smart people in one room, it’s no surprise that some of their comments trended on social media. Popular quotes:

  • “If you know where you’re going, you won’t really find anything interesting.” — Michael Levitt, 2013 Chemistry Nobel Laureate
  • “To understand others is intelligence; to understand oneself is wisdom.” — Helga Nowotny, Professor Emerita at the Swiss Federal Institute of Technology

One of the official events of the Nobel Week program that occurs in Sweden every year, the Nobel Week Dialogue is intended to stimulate public discussion on a current, science-related theme. Past topics have included genetics, energy, and aging. “This year’s Dialogue had an extremely relevant topic with fantastic speakers, a very energetic crowd, and good digital interaction,” said Nobel Media CEO Mattias Fyrenius.

Bringing together thought leaders across technology, science, economics, and other areas resonated with the live audience and social media participants, providing much needed perspective, according to Grosz. “When advances in science lead to technologies with potential to significantly affect society, that sort of scientifically solid, nuanced voice is hard to find among the clamor of media hype, which tends to exaggerate both the positive and the negative potential.”

Moderator Max Tegmark, the noted cosmologist, summed up the mood of the day by reminding participants that humans are in control. “We shouldn’t think of the future as something that will happen to us. We are the masters of our own destiny,” said Tegmark. “We should figure out what future we want and build that one.”