This article was originally featured in CIO Dive.
When we envision the future of human-technology interaction, it’s often glazed with ideas of automation and artificial intelligence (AI). After all, haven’t we all longed for the day that Michael Knight’s K.I.T.T. could be parked in our garage, allowing us to read a book while we drive down the road?
Most experts tend to agree that we’re far from engineering true AI — that is, systems that can independently process, reason and create in the same capacity as the human brain.
But here’s the question: Is replicating human intelligence the most impactful application, and therefore the key goal, of intelligence technology for businesses?
This is where the concept of intelligence augmentation (IA) comes into play. IA is the use of technology to supplement and support human intelligence, with humans remaining at the center of the decision making process.
While the underlying technologies powering AI and IA are the same, the goals and applications are fundamentally different: AI aims to create systems that run without humans, whereas IA aims to create systems that make humans better.
To be clear, this is not a separate category of technology, but simply a different way of thinking about its purpose. Arguably, many AI-branded technologies currently available for businesses can and should be more accurately be described as IA.
Consider how financial institutions integrate IA in fraud detection. Using machine learning, systems can be trained to identify and flag the markers and patterns of fraudulent activity.
Employees then use this machine data, and apply their knowledge, judgment and expertise to interpret the data, investigate and make a final call. Aside from saving time, this can also save serious amounts of money — at least $12 billion annually, according to a 2016 Oakhall study.
How can you introduce IA safely into your environment?
Here are three things to consider:
Shiny object syndrome is real — focus on the data
Business technology decision makers likely have vendors emailing them every day about exciting new AI solutions for business. Sure, robot assistants and chatbots are exciting and new, but are they really going to move the needle?
Instead, identify where massive data analysis and insights can help teams make better decisions and create higher levels of engagement, and seek out those specific solutions.
Stay ahead of change management
The big fear with introducing an IA solution is what it will mean for someone’s job. And let’s face it; for the past few decades, many have spoken about the “rise of the machines” and how humans will not have a place in the workforce.
While some entry level jobs may give way to technology, the need for people who understand data and can use analysis to make smarter decisions will be ever increasing.
Employees need to understand how IA solutions will help them do their jobs better. Also, give them opportunities to learn as part of their career growth, including certifications on specific IA tools that will be a win-win.
Experiment, but know what success is and what to measure
There’s nothing wrong with experimenting with a few different solutions to improve different workflows.
However, once those areas are identified, clear and consistent metrics are required to objectively measure the impact of IA solutions.
Without human intervention, AI solutions can introduce significant risk to businesses, as exampled by Facebook trying to get rid of their human editors.
For tasks that do not require context and are easily repeatable, AI solutions can increase efficiency without introducing risk.
But even in those cases, an under-reliance on humans, as Elon Musk recently discovered, can have disastrous results. In the near-term, the most business-relevant use cases for AI will be found in developing technology that blends the processing power of machines with human social and emotional intelligence to augment our capabilities in new and exciting ways.