Artificial Intelligence And Machine Learning
Carter also sees partner openings in artificial intelligence and machine learning (AI/ML).
“Your approach to AI/ML could be very interesting if you start with the data,” Carter said. “Before you define a machine-learning strategy, you've got to have good data that goes in because that's how you train the machine-learning model. Garbage in, garbage out. The computer can only be as smart as what you put into it.”
Carter said there's a big opportunity for partners to have a thought leadership position by telling customers, “OK, I know you want to get to ML, but let's look at the data first. Do you have the right data quality? Do you have the right data quantity? Do you have the right data source?”
She cited an AWS experiment with a university that involved 11 million lines of romance novel prose and leveraged machine-learning models.
“We trained it: 'Here's your input. Here's your data source. Here's your data quantity,’” she said. “And then we showed it a picture of two people sumo wrestling, and we said, 'OK, tell us what's happening.' So what did the computer spit out? It was ‘two men approach each other and softly kiss each other on the shoulder.’ That's not what they were doing at all. They're getting ready to wrestle.” Carter uses that exercise as an example for customers.
“They laugh, but then I'm like, 'Well, that's what you're doing here, right? You don't have the right data,’” she said. “They're so excited about machine learning, and they forget that there's a prereq [prerequisite] to making it happen.”
“If a partner for 2020 really can work on the data strategy—building the data lake and the data warehouse and helping a customer define their data strategy and executing on that—I think they're going to be the long-term winners,” Carter said. “Even though you might see some partners who go in there and do a quick-and-dirty machine-learning application, the long-term winners will be those who get the data right and then move on to the machine learning.”
It all comes back to AI and the ability for partners to deliver a “more intelligent view of a customer's data state,” according to Gavriella Schuster, Microsoft corporate vice president and channel chief.
“[It] them helps them make better decisions, which really engenders a different buyer for our partners than they've traditionally had,” she said. “It gives them a different place within the supply chain … to really add value back to the business.”