Executive Interview: Amir Banifatemim, Prize Lead, IBM Watson AI XPRIZE


“The question that everyone needs to ask themselves is, while we are building new innovative approaches with AI, how could we also think about bringing value to society?” – Amir Banifatemi

Amir Banifatemi is the Prize Lead of the IBM Watson AI XPRIZE. Amir has more than 25 years of experience in development and growth of emerging and transformative technologies. XPRIZE is a global leader in designing and implementing innovative competition models that aim to solve the world’s greatest challenges and to encourage technological development to benefit humanity.

Q. How would you describe the big picture mission of XPRIZE?

XPRIZE, basically, is an innovation engine. We find opportunities for radical transformation of society through incentivized participation of the crowd. So we leverage exponential technologies, which are the most advanced technologies, deep technologies such as artificial intelligence, blockchain, quantum computing, genetic engineering, IoT sensors, 3D printing and so on. We seek to enable breakthroughs and exponential progress of those. We are preparing society for the future by trying to identify leveraged ways to bring radical innovation to everyone.

Q. Is that what the leverage is, the prize?

Many factors come into play when you talk to teams about motivation for being a part of the XPRIZE competition. Teams say that the ecosystem, resources available and competitive spirit are front of mind – all of these drive investment into the challenge area. In many cases, the prize does factor into the equation. In certain areas, for instance, government spending might not be enough or there is no incentive for governments to spend resources, or when venture capital is not investing into certain areas, or research has not been progressing. One way to look at an XPRIZE is an opportunity to accelerate and find teams to work on those topics that have not made progress or are not going to make progress in a reasonable amount of time.

The same thing that happened about almost 100 years ago, when the first commercial aviation began. A prize of $25,000 incentivized people to travel by air from New York to Paris. Charles Lindbergh won that prize and that opened up the whole aviation industry. The number of people trying to be creative and innovative about aviation — engines, everything — went through the roof so that opened up the whole aviation industry at that time.

And the same way, today, we can say that in 2003 with SpaceShipOne winning the $10 million prize to incentivize engineers to create a craft that would transport a three-person crew 100 kilometers into the atmosphere. That was really the beginning of commercial space travel. Virgin Galactic purchased the license of that and started Virgin Galactic. A number of startups and research then spun out and the whole space industry opened up. And, today, you see how many people are trying to go to space, to go to Mars, to explore other planets. That was basically the opening.

So what we’re saying about leverage is that when we open up an opportunity to a number of teams with a very strong innovation challenge and people respond positively to that challenge, then a number of innovations happen and unlikely teams get created. Because many, many people may have opportunities and ideas about solving a problem who may not come forward, necessarily. So we create an incentive.

Q. XPRIZE in December announced 59 teams advancing in the $5 million dollar competition for the Watson IBM Watson AI XPRIZE, which is a four-year global competition. Where are you now on that schedule?

The global competition started in June of 2016. We invited teams to come and compete for that prize, sponsored by IBM. Meaning, that the purse is given by IBM for the winners, therefore the name of IBM Watson on the prize. And at that time, about 10,000 requests came through; 800 teams got started. Out of those 800 teams, 150 teams got an approval by the judges to start the competition because they demonstrated the needed capabilities. The competition is a four-year competition with different milestones.

The first milestone is year one, where we go from round one to round two. And to go to round two, teams have to demonstrate that they’re working on meaningful projects and they are tackling the guidelines of the competition. So 59 teams out of the 150 demonstrated that, so these 59 that we announced in December. Now we are going into the second year, engaging in solution development. At the end of 2018, the teams will submit their work for judges to review and decide who is going to round three. Round four will be the semi-final; the finalists will be on the TED stage in 2020. So this stepping stone to a prize is usually the way an XPRIZE functions. It’s always a longer process. It’s not a three- or six-month effort. And, today, teams are gradually engaging into their solution development, now.

Q. In your release, you announced the top 10 teams. Could you comment on what you’re seeing from these AI innovators?

Yes. It doesn’t mean that these teams are going to win. It means that at this phase, on round one, these are the teams that demonstrated more advancement and have put more work into the competition. The domains in which they are working include health and wellness, civil society, space and frontiers, learning and human potential, shelter, energy, and planet and environment.

So these teams are actually working on very specific problems related, for instance, to managing depression, or helping with democracy, or helping with improving the environment and clean water. Some of the 59 teams are tackling problems in the same domain, but they definitely have different focuses and the type of AI and technology they’re using is very varied. We may not have full knowledge of what they’re doing yet. We have validated that they’re working on something meaningful, which has impact attached to it and they have demonstrated that they know what they’re talking about and they have the technical abilities to deliver. What they’re actually going to be delivering, we will see in the next coming months because they’re actually building it.

Q. How do you characterize the entrants? Are they practitioners in the field who are working in business? Or are they students or both? Or is all types?

We have actually a good mix of teams created with individuals coming from academia, from research, from corporations; most of them are startups. There are a few students but not many. Teams have to be multidisciplinary because to tackle AI for impact, you have to tackle both — AI and impact. So if you’re talking about a project such as improving enterprise quality control, for instance, you need to know something about quality control and you need to know something about AI. So for that reason, teams are usually multidisciplinary, have both sides, the AI side and the AI/machine learning and everything that’s related to AI but also the domain, the expertise of the domain. So those teams are not just teams of students, or just a team of academics, or just a team of corporate people, or just a team of business people. These are very much mixed teams.

Q. What is the best use of AI for the most impact?

I’m trying to bring attention to the fact that AI can used for the most immediate and pressing issues that society faces. Whether that issue is unemployment or economic output or healthcare or education. And corporations have to be imaginative and modern in creating new products and more wealth and more jobs. Every now and then, some technologies come forward and give us the opportunity to think better about applications built with them. Machine learning and artificial intelligence give us more power to predict, to analyze, to understand, and to create more automation and to learn more with data. So we ask how can we employ these capabilities to problems that are definitely more important to tackle, such as for everyone to have a better life, to have better welfare, to participate more in democracy and to basically have better skills and to be employable and so forth. This is how we look at impact.

The question that everyone needs to ask themselves is, while we are building new innovative approaches with AI, how could we also think about bringing value to society? So think of it as a corporation that makes money and profits and generates new value, and is also thinking about being sustainable, having less impact on the environment and helping better its employees and its community. So AI gives us the opportunity to ask a number of questions. Of course, artificial intelligence is built with algorithms, computing power, data. Most of the time it can be autonomous. Sometimes it cannot be autonomous, and it has to be used with human help. The amount of information that we inject into AI programs certainly contains our biases, and certainly exposes people to less privacy. How can we also be aware of those? So that question about impact brings a lot of other questions next to it, which I’m suggesting we should be aware of and think about. Corporations have a role, definitely, to participate in this dialogue. And while they’re pursuing reinvention of their businesses with AI, I think there’s an opportunity, also, for them to understand the implications of AI through the whole company itself and the ecosystem in which they operate.

Q. What is the impact of AI on corporations reinventing themselves?

Today AI is set to transform all business in ways that we have not seen since the industrial revolution. So, really, we are in a new revolution. And, fundamentally, AI is helping business reinvent how they run, how they operate, how they compete, how they thrive, how they create value. So if, for instance, technologies help  to lower costs and create new jobs or create new growth opportunities — this is how AI is basically helping enterprises reinvent themselves.

Maybe the type of jobs and skills they need to incorporate is going to be different. Maybe the training of employees has to happen sooner, maybe lowering costs and increasing productivity helps us generate better profits that could be repurposed to other areas. Maybe some aspect of jobs will be lost and then can be converted into something else. This can help not only enterprises, but it also helps governments, nonprofits, and the society as a whole. And I think the understanding of that topic is critical. So the impact of AI on corporations is really the opportunity to think again about, one, how can we create value? And second, how can the tools of productivity and growth play out, again, with AI as an ingredient?

Many people say that AI will boost profit and innovation. Some people say that AI will lead responsiveness. Some people think that AI is going to bring more inclusion. Some people think that AI will increase automation so that jobs will be lost. All of these are interesting to consider. The impact is going to be central to each organization to decide upon and take action. And I think that opportunity, again, the same way it was important for society, it’s important for enterprises because it’s part of the core of creating jobs and value and wealth. And because products and service are fundamental to the growth of enterprises, AI is definitely a conversation to have. I think all big knowledge partners, consultancies — everyone basically says the same thing about AI helping enterprises reinvent themselves today.

Q. What would you say to people who fear AI?

For many people, AI is unknown. Many uses of AI are flying somewhat under the radar for most people. We’ve seen AI used for dramatic effect, and for some, the knowledge of AI comes, mostly, from science fiction and media portraying AI always in the form of a Terminator or dystopian face or dystopian scenario. AI in the real world is different. AI is, as a core, is a set of technologies that help us automate certain things. Automation will probably make our life better. Automation will help us do things stronger, better, faster, in a larger scale.

And any new tool probably creates a number of fears and the fear of unemployment is probably the number one fear today. However AI is allowing us to create new jobs as well. The fear of losing privacy and being observed is a second fear. I personally don’t subscribe to those fears as long as, we, as a society, as a group, as a collective, take actions to make sure that AI is built in a beneficial way.

AI creates new ways for us to reboot everything and think about it and talk about it. But I don’t transfer that into fear. I transfer that into a responsibility that we have to ask those hard questions and make sure that we have good boundaries around that and good checks and balances and make sure that AI is, first, beneficial and built safely. And, second, that we use AI for the right reasons and we don’t put it everywhere and that we have some responsibility and some third-party check-in’s and so forth.

Q. Are there key areas of AI where you’re seeing innovation today?

AI is advancing very fast in certain areas, such as healthcare, which has seen tremendous benefits. Imagine cognition disease diagnostics, assisting doctors with robotics, assisting with some medical therapeutic acts and, basically, planning in general of healthcare issues. But, also, in education, where personalized learning is becoming more important. You can have more interactions, you can incorporate automation, in terms of certain types of experiences in learning. And AI is helping with climate and weather and water management, so climate and environment is getting benefits from it. Of course, everyone sees self-driving cars as automation happening on the road, where we can now detect objects and manage the car movement at a certain level. Today, we’re not at full capacity of level five autonomous driving but we are close to level three. And teams and labs and corporations are working on improving that. These are the low-hanging areas where we see immediate application but corporations, government, everyone is looking at AI in everything — from economic participation, to democracy, to helping psychology issues or helping committees connect better — we’ve seen so many examples just with this competition. But this competition represents only a tiny, a very tiny fraction of all the ideas and possibilities that many people are working on today.

Q. What message should readers of AI Trends take away?

Well first, and foremost, I think it’s important that everyone one of us understands better what AI is and is not. And understand what it can do and try to participate in dialogue and conversations around privacy, ethics and governance of AI. Test products, give feedback, and participate with groups and teams that are working on hard topics and are trying to collaborate and collectively make sure that AI is well-understood. It’s a real change and I think participating, understanding, and discussing it and sharing the right facts and data is important. Today we have opportunities to showcase practical applications that benefit everyone. Let’s identify those and push those first to be implemented, for the majority of people to benefit and to create a better society. So I will say this is the first stage that we can focus on and we have a responsibility to focus on it.

For more information, go to XPRIZE.org.