Excitement or caution: where do you sit on this spectrum regarding the media mania regarding AI?
I’m offering you a false choice, of course: the two positions are not mutually exclusive.
Creating a better future for humans – enabled by technology
The thing that most excites me about the near future of AI is the way it is creating something like a human/machine society, with potential for new possibilities and improved services. Whether this development leads to a transhuman society or not, that I don’t know, but I am quite excited about its possibility.
Can anybody honestly declare there is no risk from AI in the future? No, they can’t. And that means we must exercise caution and foresight in pursuing our AI objectives.
There are plenty of reasons to see AI as being positive for society. I break these down into visible and invisible developments.
On the visible side, we can expect to see AI producing significant changes in the way services, such as healthcare, are delivered, with service providers – doctors, consultants, diagnostic specialists – gaining powerful partners in the form of AI collaborators. For example, the very rapid and precise diagnosis support from AIs that we are already seeing in areas such as radiography will soon be available in personalized medicine – a domain too complex for routine diagnosis by humans – enabling tremendous benefits to ensue.
There has already been progress in pharmacogenomics, which recognizes that drugs do not work uniformly across an entire population but vary according to individuals’ genetic make-up. AI provides the diagnostic firepower to make real progress here.
Likewise, in retail and finance. Here, highly personalized, AI-supported services will become the norm. Call center service will be not only be provided by intelligent bots but created by AI bots. AI is already leading to the emergence of more personalized user experiences, dynamically adapted to recognize context, location and preferences.
In the consumer market, digital virtual assistants such as Microsoft Cortana and Amazon Alexa are already familiar. Adoption is starting to expand rapidly in the enterprise world: According to a recent whitepaper and survey, Workplace 2025, created for Fujitsu by leading independent research firm Pierre Audoin Consultants (PAC), close to half (47 percent) of organizations plan to invest in this area in the next two years.
Uptake will be particularly strong in financial services (64 percent), with banks including Sweden’s Swedbank and India’s HDFC Bank already using AI-powered virtual assistants in customer services roles. And the AIs we work with day to day are getting better and more naturally interactive. Chat bots will soon be indistinguishable for human agents and we will find ourselves working very effectively in partnership with AIs in all aspects of our lives.
The invisible impact of AI
However, the industry segment with the possibly largest but mostly invisible effect will be manufacturing.
AI and manufacturing automation open doors to new products, delivered in different ways. At Fujitsu’s own manufacturing facility in Augsburg, Germany, computer mainboards are now manufactured and tested with robots and humans working side by side. Robots handle the monotonous and repetitive tasks after a human worker shows the robot the required assembly steps, which are then embedded in the robot’s capabilities using AI.
The impact of AI in manufacturing goes well beyond single machines or production lines, however. The insights that it allows into how factories or entire manufacturing value chains are performing, including raw material delivery at one end through to retail or home distribution at the other, will result not just in higher levels of manufacturing efficiency but the ability to predict maintenance issues and intervene before they arise. This will all be taking place at a level where many of us will never see it.
From these examples you can see that the way we work is going to change. This realization has resulted in more press coverage than necessary about two aspects of AI.
First is the fear that AI will take away people’s jobs and livelihoods. While AI will certainly change the labor landscape, it will not necessarily be all in one direction. Just like many other major advances in technology, I believe that AI will create new roles in society as well as changing existing ones. We don’t yet know the net-net effect, therefore preparation for all eventualities is needed.
The other issue is the scare-mongering around the possibility that AIs will take over and eliminate humans. While the scenario is possible, it is far, far out in the future. What we can do at this point is to start a sober societal debate of how we can prepare.
For example, if we’re thinking about the changes to the delivery of healthcare and banking that I mentioned earlier, then building the right kind of trust in the AI services of the future will be key. In some sectors, AI will need to be able to justify the decisions and conclusions it has arrived at, in the same way a banker can tell you why you were turned down for a loan today. This will become even more critical as large parts of future AI infrastructure begin to be created by AIs themselves.
In the long run, AIs will build their own trust through our familiarity with them, but this will take time as it has more to do with human factors than technology. Until then, we need to walk the fine line between excitement and caution.