AIBusiness.org has secured an exclusive interview with Gideon Mann, Head of Data Science for Bloomberg L.P. Gideon shares how one of the world’s leading media brands is planning on building hybrid systems that augment human ability. AI will help Bloomberg deliver a significant edge to clients for being first, better informed and from taking risks.
Q: Bloomberg has been working on a number of different projects in AI. What innovations can the business world look forward to?
A: At the core, Bloomberg’s mission is to provide our clients the critical data, news and financial insights they need to make fast, smart decisions. We do this by increasing the transparency in the market so that it’s not who you know but what you know. The biggest challenge our clients face is dealing with the volume, velocity and variety of relevant data. We apply AI to help them make sense of the vast amounts of business and financial information available to them. All of our efforts in areas like natural language search, Twitter sentiment analysis, machine learning search ranking or targeted news alerting, are aimed at increasing client capacity to make the best possible decisions. As a result, you can expect to see both product and overall technological enhancements and advancements from us.
Q: Bloomberg has unrivalled experience working with leading financialorganizations and is trusted by the enterprise community globally. Which markets and regions do you see leading the way in relation to enterprise adoption among your client base?
A: In the financial industry, our global client base gets an edge from being first, better informed and from taking risks. Everyone is seeking out new ways to operate, and I’d suspect the best are moving the fastest no matter their geography or industry.
Q: What is the current roadmap for Bloomberg’s own solutions under the umbrella of AI and how does it fit under your broader proposition?
A: Our rough roadmap is to build hybrid systems that augment human ability with sophisticated computation, transitioning from merely making financial information available, but helping our clients cope with the vast amount of information available. It is now common practice to make quick decisions on the basis of earnings reports, but the window of opportunity to trade on new detailed information that is available in filings and in news is still open.
Gideon Mann, Head of Data Science for Bloomberg L.P.
Q: AI is changing – accelerating – the pace of productivity in the workplace. What are the biggest opportunities for an enterprise looking to implement this level of software?
A: I’d look for human decisions that are repeated over and over, and have a high level of agreement on what the correct answer should be. With enough data on the correct choice, it’s increasingly easy to build custom AI solutions.
Q: For large enterprises, which technologies under the AI umbrella do you believe will have the biggest impact across all verticals?
A: Recently, advances in computer vision have been staggering – and they enable a whole suite of products. Object recognition and face recognition are at the point where they are almost a commodity, and that simply has never been true before. I would contrast this with progress in natural language processing, where we are still far from commodity solutions.
Q: Who will ultimately be driving AI adoption in an organization – is it the CIO, CTO, CDO and/or directly Heads of departments/divisions?
A: Typically, direct heads are able to recognize the opportunity for AI, but often they don’t have existing resources to capitalize on the opportunity, and so they reach out to internal champions. I would suspect every organization is somewhat different, but at Bloomberg, when engineering groups are looking into a new opportunity, key tech leaders in the CTO department are sometimes the first people to be contacted when engineering groups and other times sometimes champions come from inside R&D itself.
Q: How do you see the market demand for AI applications evolving in the next 5 years?
A: Overall, I’d expect demand will only increase as AI driven applications provide cheaper and better services. I think it will also be interesting to see changes in the marketplace for AI –enabling technology for companies looking to implement machine learning: what can be acquired in B2B as a commodity and what has to be done in-house. The first wave of commodity Prediction APIs were not successful because they didn’t add enough value over what open source machine learning libraries could provide and introduced unpleasant vendor lock-in. This suggests that value from marketplace machine learning must come from higher up the value chain like object recognition, for example.
Q: Looking more broadly ahead, what do you think a technologically up-to-date future workplace will look like 5 years from now?
A: The increased ability for AI to augment routine human actions means that there will be less time spent on the process of work – e.g. paperwork – and more time spent in things like problem solving and ideation. Administrative jobs as we currently know them will need to evolve. I expect things like creativity, story-telling and communication – things uniquely human — will take on an even greater importance throughout organizations. Oddly enough, this is likely going to emphasize interpersonal skills, like empathy, that sometimes get short shrift when building the technology itself.