It’s 1997. The Backstreet Boys debut in the U.S. with one of the most successful albums of all time. Microsoft is the world’s most valuable company, with a $261 billion market cap. And an IBM computer named Deep Blue defeats Garry Kasparov, reigning world chess champion and, at the time, the highest-ranked chess player to have ever lived.
Though it was not the first time man has lost to machine, it is perhaps the most prominent, highly publicized by IBM and widely covered by the global media. It was viewed as a milestone for AI, the true arrival of computer intelligence. The world celebrated the achievement of technology — or offered doomsday predictions of a robot revolution. The future was now.
It should sound familiar. Google’s DeepMind AI recently won a landmark victory over Go world champion Lee Se-dol. Go, a board game substantially more complex than chess, has long been considered a penultimate challenge for AI. Where brute-force calculations give computers an edge in chess, Go has a near-infinite number of possible board positions. Computer power was not enough to defeat human intuition — until now.
The fact that AI has been able to operate at this level is being viewed as a sign of the maturation of AI, and we see many of the same themes discussed today as were prevalent around Kasparov’s defeat nearly 20 years ago. Then and now, the hype and hysteria misses the forest for the trees. While these achievements are notable, we are still many years away from creating the true, autonomous artificial intelligence that will be able to significantly change our lives.
Humans being beaten or replaced by computers is a much more popular theme than computers making humans better. Deep Blue and DeepMind’s respective victories highlight how far AI has come, but progress in this space shouldn’t be measured in wins and losses. It is actually far more exciting to look at AI’s development in context of how it has empowered humans through IA — intelligence augmentation — to change the world in more meaningful, immediate ways.
One of the most interesting consequences from Kasparov’s Deep Blue match was the resulting trend in professional chess players embracing the use of machines. Kasparov was one of the first top players to incorporate them into his training regime. Rather than reject computers, Kasparov realized he could combine the long-term strategic thinking of the human with the computing power of the machine to improve his own play.
The computer, proven capable of solving chess problems, would now be leveraged to help humans solve more complex problems neither could handle on their own. Kasparov adapted to this new reality much faster than any of his rivals, enabling him to retain the World No. 1 ranking for 29 years, an incredible feat that no other player has come close to achieving.
AI is improving rapidly, but we are still far away from creating human-level intelligence in a computer.
This is the most basic illustration of the development of AI for practical applications. AI will become capable of handling one specific task, but truly impactful results don’t come until humans re-enter the equation and apply context and direction. The days when a computer intelligence will be able to autonomously run a business or act as your effective assistant are still far off. But IA is real and directly impacting businesses and consumers around the world today.
IBM’s Watson can be considered the world’s most popular AI, with mainstream global recognition following its Jeopardy! victory. Yet far more important than its status as AI that is great at trivia is its ability to help doctors improve their diagnoses and save lives. Today, Watson powers a large range of IA use-cases across healthcare, directly improving how doctors do their jobs.
For example, while images encompass as much as 90 percent of all medical data, it can be difficult for doctors to extract vital information from them. Watson’s technology can process massive amounts of medical images, giving doctors the resources to make smarter medical decisions and pick up symptoms that would otherwise go undetected.
In other areas, doctors are using Watson-based technology to provide more accurate and personalized information in response to patient queries, match therapists and patients in real time based on emotional indicators and more.
Many point to digital assistants as proof-positive for AI’s potential to improve human life. Indeed, this is one of the most exciting areas for computer intelligence — yet the most powerful use cases come from IA rather than true AI. Siri’s utility, for instance, is still somewhat limited. Apple’s program is AI in that it employs audio recognition and machine learning to deliver information, yet it is not capable of connecting the dots to a decision. Siri enables humans to find information more easily, but it is the human’s intelligence that ultimately acts.
Facebook’s M is a perfect example of IA acting as the bridge to AI. M has initially been far more able to not only find, but act on, information — yet it quickly became apparent that there was ultimately still a human on the other end. M is a powerful IA use case; a consumer feeds a question or problem into software, the software breaks down the query and organizes relevant information and a Facebook employee digests and acts on the information in a very quick timescale.
Automobiles are another area people expect AI to transform, with computer vision powering a new era of automation and self-driving cars. While great strides are being made in this space, few are ready to fully trust their lives to an AI-controlled car. However, companies like Tesla are experimenting with ways to use IA to make this dream a reality today. Tesla Autopilot uses AI for basic transportation and collision detection, but still puts a human being at the wheel to make decisions such as when to change lanes or take over when the software gets confused.
AI is improving rapidly, but we are still far away from creating human-level intelligence in a computer.
We should celebrate the landmark moments of artificial intelligence, but we need to remember the true value of having a computer with the ability to beat us at chess, Jeopardy! or Go. These moments don’t show us the future, they don’t show us how AI will change our lives “one day.” Their power is in influencing us to rely on computers more and integrate them into our daily lives. IA is our bridge to the future. One day, computers may replace humans at many jobs. For now, they are making us better. That is worth celebrating.