A few weeks ago, Jochen Leidner, Director of Research at Thomson Reuters, discussed and dissected artificial intelligence (AI). Leidner offered some case studies, conducted by his team, to demonstrate successful ways of building AI products.
However, before getting into specifics, Leidner gave a very brief explanation of what AI is and what it does.
“It automates things that we deemed, maybe, not automatable a couple of years ago – if it works, or to an extent.”
Dr. Leidner, who is also the Royal Academy of Engineering Visiting Professor of Data Analytics at the University of Sheffield, made the point very early on that AI is a tool to be utilized with other forms of intelligence.
He explained that Thomson Reuters uses AI technology along with several other trusted content sources to provide the most accurate answers.
Thomson Reuters has partnered up with Barclays to help support the start-up community and encourage innovation, by participating in a series of seminars at Barclays Eagle Labs in Cambridge, a workspace for local start-ups.
“At Thomson Reuters, we have a deep appreciation for human domain expertise… Together with content sources and the technology, it’s really this mix that provides the value. I think it is important to understand that this together provides the actionable intelligence.”
Leidner then posed a question to the audience: “What is AI?”
Watch video — How do you build smarter AI products?
Where’s the Intelligence in Artificial Intelligence?
The audience was right in-step with Leidner when they said that artificial intelligence, in the literal sense, does not yet exist.
Leidner said, “No system ever asks a question unexpectedly in the way children do, this is kind of my measure when talking about things as intelligent systems – no longer predictable but somehow insightful.”
Though the intelligence in artificial intelligence may have yet to be truly developed, companies are still investing billions in AI products and entire industries are making it a priority.
According to a study published this summer from the McKinsey Global Institute:
- Tech giants including Baidu and Google spent between $20 to $30 billion on AI in 2016, with 90% spent on R&D and deployment, and 10% on AI acquisitions.
- The total annual investment in AI was between $8 and $12 billion in 2016, with machine learning attracting nearly 60% of that investment.
- High Tech, Communications and Financial Services are predicted to be the leading industries to adopt AI in the next three years.
However, Professor Leidner suggests this may just be another cycle of AI hype.
“AI has been pursued for a long time, there were several hype waves and several, so-called, ‘AI winters’ where funding for artificial intelligence research dried up because the previous hype did not deliver. I wouldn’t be surprised if we’re moving into another such cycle.”
Watch video — How will AI reshape the legal landscape?
A Pragmatic Definition of Artificial Intelligence
So where does that leave us? Should the term AI be forgotten? Should companies give up on this expensive idea and chase for artificial intelligence?
According to Leidner, no. He was quick to add that these hype cycles don’t matter much in the long run because each has spurred new insights and new developments.
In other words, while we may not have achieved real artificial intelligence, we are still creating technological advancements that make things automated, easier and more efficient.
The focus always ought to be to focus one’s thinking on the problem, the solution and the measurement of the solution’s quality in quantitative ways.
Leidner suggests we become more pragmatic in our research and discussion of AI by taking on a more realistic definition:
“Systems that seem smarter and are actually smarter than before – smarter than the old systems rather than smarter than humans – and complementing humans to make a hyper-productive team of humans supported by machines, not humans replaced poorly by machines.”