Two years ago, billionaire tech venture capitalist Vinod Khosla made the bold prediction that in the next 20 years, “machine learning will have more impact than mobile.” He’s still got 18 years to go, but his prediction is already coming true.
Machine learning is the underpinning for many common learning technologies, including learning management systems, MOOCs, data analytics and other tools; and many learning vendors have marked it firmly on their future road maps. The allure of machine learning is clear, said James Cross, director of learning product strategy for Workday. “Research shows that it increases engagement and makes the learning experience more relevant and ‘sticky.’ ”
The technology works through a series of algorithms, or computer-based queries that allow a piece of software to learn from data over time so it can identify trends and patterns that inform future searches and suggestions. In other words, the software watches what learners do, and how they engage with content, then proactively provides them with content they need the moment they need it.
For example, if a learner always chooses audio content, or skips over lessons on a certain topic, the machine can learn from that, and make more informed suggestions in the future. Similarly, if the machine sees that specific types of employees choose the same set of courses, it can automatically recommend that content to new employees in those roles.
“The ultimate goal of this technology is to create a more custom learner experience,” Cross explained. “Think of it as Netflix for learning.”
In addition to telling employees what training they need, machine learning can save them from having to complete courses they don’t need, said Joshua Crumbaugh, CEO of PeopleSec, a cybersecurity training company. PeopleSec uses machine learning and artificial intelligence in its computer security awareness training to reward employees for good behavior, and to increase training requirements for those who are higher risk. “We can be more invasive with employees who need it the most, and less invasive with the rest of the staff.”
Such customization cuts training time and cost because it’s only required for those who need it, and is respectful of the learners’ time. And because employees know training is linked to whether they embrace the lessons learned, they have more incentive to change their behavior. “If they don’t want to end up on the negative list, they know they shouldn’t click on phishing scams,” Crumbaugh said.
There are also social implications for machine learning in the workplace, said John Schneider, vice president of product marketing for Jive Software. When users rate a course, or participate in an online discussion about a topic, machine learning tools can capture and learn from those interactions. It can indicate which courses are successful, where knowledge is falling short, and who are the influencers and subject matter experts for a given topic, he said. This can be powerful information for CLOs. “When they can tap into the knowledge of the company, it strengthens their ability to chart the learning curriculum.”
—
Machine learning offers exciting opportunities for learning leaders to deliver better, faster, custom content. It also should prompt them to take stock of their current skill set. CLOs in a digital world may not need an IT degree to be competitive, but they do need to understand the technology behind the tools they use. Then they can make the right investment decisions, and use those technologies for strategic business benefit, said Ian Barkin, cofounder of Symphony Ventures, a future of work technology consulting firm. Machine learning enabled tools can automate common tasks, like updating LMS databases, transcribing documents or assigning content. “That can free CLOs to focus on more important tasks, like culture building and strategic planning,” he said.
But he said none of these advances are plug and play. Learning leaders need to add data analytics talent to their teams if they want to make the most of these tools. Consider IBM Watson, one of the most famous machine learning computers in world. Watson was built by IBM researchers to answer questions posed in natural language, and went on to win the million-dollar prize on the game show “Jeopardy.” “Watson didn’t just magically work,” Barkin said. “It took years of training and programming to teach it how to think.”
As with all new technology, that’s where the real work comes in. Before a company can make use of machine learning or artificial intelligence to improve their learning strategy and programming, they need to figure out how the technology applies to their business case, what benefits they want to derive from the tools, and how to make that happen, said Dani Johnson, vice president of learning and development research for Bersin by Deloitte. To do that, learning leaders need to educate themselves on how these technologies work, what questions they want answered, and how to use the information they glean.
They also should think about how these tools will change the way they do their jobs and the knowledge their people need to acquire. Johnson said in the past two years there has been a huge shift in the skills organizations need. The most innovative CLOs are crafting new learning strategies in response to technology trends to meet long-term talent development goals and to effectively engage learners. “There is a huge opportunity here for forward-thinking CLOs to have a dramatic impact on organizational strategy — if they are willing to adapt.”
Sarah Fister Gale is a writer based in Chicago. Comment below, or email editor@CLOmedia.com.
