We live in a time of great transformation. Over the last thirty years, technology has fueled a host of changes in how we live, work, and enable the public good.
Now, we’re on the cusp of a major change that could revolutionize almost every sector, industry, and government service.
This change is Artificial Intelligence (AI) — and there are three key capabilities that separate AI from many of the past technical or scientific breakthroughs in terms of the value it can create.
When people usually think of AI, it’s in the context of machine learning and the ability to process large amounts of data to connect the dots across thousands of variables. Even the smartest humans have only such ability to consume large amounts of information and to process a handful of variables. This is where machine learning comes into play. Machines have the capability to process, track, and draw insight from millions of data points very quickly. That’s why healthcare researchers can leverage AI to develop new target proteins in the fight against cancer in just a matter of weeks rather than months.
Likewise, lawyers are using AI tools to reduce the amount of time to conduct legal research and increase the time they have for analysis and case development. However, the machine learning component is just one of three key capabilities that enable AI solutions to bring value.
The second key capability is natural language processing. When people talk, a lot of information is communicated contextually and non-verbally. Moreover, think about all the slang, idiom, and jargon we use in our normal, everyday conversations. If someone were to say, “I’m feeling blue because it’s raining cats and dogs,” most people would understand the intent. Now, consider if a machine heard this, what would it decipher? The person is the colour blue because dogs and cats are falling from the sky? Most machine solutions think this way because they leverage keywords to draw meaning. However, AI solutions like IBM Watson draw context from the statement by looking at the grammar, word choice, tone of voice, and place within the conversation. Thus, Watson would understand that the person feels sad because it’s raining so heavily.
The third key capability is the interactive capability of AI solutions. We’re shifting away from having to define precise requirements and scenarios to enabling conversations as the drivers between human-computer interactions. This is a major change. Consider a person who wants to buy a bicycle.
How would they figure it out today? They could do an Internet search, visit forums, talk to current bicycle owners, etc. to gather information and make a decision. With AI, this is a conversation. Imagine an AI solution that’s an expert in bicycles and knows the person both emotionally and psychologically. This person can go to the AI solution and ask, “Which bicycle should I buy?” The AI solution would engage the person by asking questions like: Why do you want a bike? How much do you want to spend? Where do you plan on riding it? Based on this dialogue, the AI solution can make a personalized recommendation for this person in a matter of minutes.
With these three key capabilities weaved together, businesses, scientists, researchers, and governments are using AI to outthink their biggest challenges.
For example, IBM is using Watson to solve some of Africa’s challenges in agriculture, health care, education, energy, and water through an initiative called Project Lucy (see infographic). Consider health care, where there is only one doctor for about every two thousand people. As a result, IBM is creating a Watson-powered solution allowing people with minimal healthcare knowledge to help diagnose and treat medical conditions. These people can talk with Watson, share information through text, audio, or images, and get immediate help from its subject matter expertise.
Similarly, Sesame Street is leveraging Watson to help advance preschool education worldwide by creating a personalized, adaptive learning environment for young children. Likewise, film studios are using Watson’s capabilities to develop movie trailers. By having Watson watch the movie, it can draw out the emotional context of each scene and determine the optimal selection and ordering of movie snippets that will entice people to go see the movie. Additionally, athletes such as Serena Williams are using Watson for training. This includes game preparation as well as conditioning based on the athlete’s playing style and medical history.
Even human-resource professionals are leveraging Watson. In recruiting, Watson’s ability to generate psychographic profiles can help determine if a job candidate will fit in with the team and the corporate culture.