Sramana Mitra: What format does that take? Is this like an online chatbot where there are three entities interfacing?
Paul Daugherty: The only people conversing are the customer service agent and the customer. The chatbot is advising the customer service agent.
Sramana Mitra: Behind the scene.
Paul Daugherty: Right. It’s allowing the person to be more effective and having more human conversation.
Sramana Mitra: Conceivably, there could also be a scenario where this is a three-way chatbot conversation where you acknowledge that there is a machine that is also providing interesting insights and data. Then the agent and customer is taking that input into the process. That could also be an interaction format.
Paul Daugherty: In this area of conversational AI for large call centers and customer service organizations, it’s probably the fastest growing area we see with artificial intelligence right now. It has huge applicability across many wireless companies. We’re seeing a lot of this work at scale right now.
A common scenario is, the customer might start with a chatbot to handle some basic transactions they have. Many organizations are deploying this capability. Then it might get to a point in the conversation where the human customer service agent might come into the mix. That’s exactly the scenario that you talked about.
Sramana Mitra: Other fields that you have good use cases in?
Paul Daugherty: We’re doing interesting work in agriculture. We’re using drones and video analytics in the agricultural process. In one case, it’s in a forestry type of environment where they’re growing new produce continuously and managing the density of the trees they’re growing and looking for underbrush or other things that might be interfering.
Before, they couldn’t accurately and completely monitor the growing process. Now they can. They can now identify issues, where they might need to space out the trees, or where they might have underbrush. We work with companies to identify drugs faster. If you’re a pharmaceutical company coming up with couple new drugs a year, which is the way the industry works, being able to come up with one more quickly is a big deal.
Sramana Mitra: Is there any example of a drug that has been discovered using AI yet?
Paul Daugherty: None that I can talk about yet. Several are at the work in progress stage. Industries where new materials are being developed are interesting areas to look at.
There’s a company called Zymergen. They’re using AI in microbial engineering. They’re creating new materials using artificial intelligence. They’re replacing petroleum-based products by using AI to identify microbes that can grow new materials with the same characteristics as materials we currently produce through petroleum. It has massive environmental benefits. It can eliminate the dependence on petroleum. You can also create materials you couldn’t create before.
Think about very flexible electronics that can wrap around your wrist in the form of jewelry but have electrical characteristics that can be used for communications. These are the kinds of material you can create from microbial engineering using AI rather than the typical materials engineering process.
This segment is part 2 in the series: Thought Leaders in Artificial Intelligence: Paul Daugherty, CTO and Chief Innovation Officer of Accenture