IBM can predict when employees are ready to quit with 95 percent accuracy, according to CEO Ginni Rometty.
Speaking at a Work Talent and Human Resources Summit in New York, Rometty described how the company's "predictive attrition" software has reportedly saved the company $300 million by identifying which of its roughly 350,000 workers are on the cusp of quitting, reports CNBC.
The way the system works is a trade secret - however a 2016 blog post described how IBM used their Watson Analytics smart data program for employee retention.
"The best time to get to an employee is before they go," she said.
IBM HR has a patent for its "predictive attrition program" which was developed with Watson to predict employee flight risk and prescribe actions for managers to engage employees. Rometty would not explain "the secret sauce" that allowed the AI to work so effectively in identifying workers about to jump (officially, IBM said the predictions are now in the 95 percent accuracy "range"). Rometty would only say that its success comes through analyzing many data points. -CNBC
"It took time to convince company management it was accurate," Rometty added.
According to the 2016 post on job attrition, factors include an employee's age, marital status, stock options, how many years they've been with the company, and the how far they've been promoted.
Rometty also said that the system can eliminate human resources jobs, which she described as needing an overhaul. Since implementing the system across their cloud services, IBM has reduced its global HR department by 30% - while remaining positions are higher paid and higher value.
"You have to put AI through everything you do," said Rometty.
Among the tasks that HR departments and corporate managers have not always proved effective at, and where AI will play a bigger role in the future, is keeping employees on a clear career path and identifying their skills.
Rometty said being transparent with individual employees about their career path is an issue in which many companies still fail. And it is going to become more critical. "I expect AI will change 100 percent of jobs in the next five to 10 years," the IBM CEO said. -CNBC
"If you have a skill that is not needed for the future and is abundant in the market and does not fit a strategy my company needs, you are not in a good square to stay inside of," said Rometty. "I really believe in being transparent about where skills are."
With transparency and enough data points, IBM's AI can zero in on an individual's strengths, enabling management to make optimal use of their workforce.
"We found manager surveys were not accurate," said Rometty. "Managers are subjective in ratings. We can infer and be more accurate from data."
Instead, AI is able to view the tasks employees are completing, what training they've had at the task, and any rankings they've earned to perform a much more accurate assessment of the employee's skill set vs. manager surveys.
Traditional human resource departments, where Rometty said companies typically "underinvest," has been divided between a self-service system, where employees are forced to be their own career managers, and a defensive system to deal with poor performers.
"We need to bring AI everywhere and get rid of the [existing] self-service system," Rometty said. IBM employees no longer need to decipher which programs will help them upskill; its AI suggests to each employee what they should be learning in order to get ahead in their career.
Poor performers, meanwhile, will not be a "problem" that is dealt with only by managers, HR, legal and finance, but by solutions groups — IBM is using "pop-up" solutions centers to assist managers in seeking better performance from their employees. She said many companies have relied on centers of excellence — specialized groups or collaborative entities created to focus on areas where there is a knowledge or skills gap within an organization or community. "We have to move away from centers of excellence to solutions centers." -CNBC
IBM's MYCA (My Career Advisor) AI virtual assistant can leverage the Watson system to help employees identify where they need to beef up their skills. It's companion, "Blue Match" technology - can find jobs for employees based on their AI-inferred skills data, according to CNBC. According to Rometty, some of the 27% of IBM workers who were reassigned or promoted in 2018 were assisted by Blue Match.
"AI will change all jobs once it is in the workflow, and that is the most meaningful kind of AI. Yes, some jobs will be replaced, but that is a red herring," Romney said. "It is about getting people to work at the intersection of this."
"This is all a game about skill and having people with the right skills, and everyone's job is changing."
Could IBM's AI have seen this coming?
(Charlo Greene was later charged with 10 felonies and four misdemeanors for selling marijuana to undercover officers in her Anchorage dispensary. She pleaded guilty in 2018 to one count of misconduct involving a controlled substance and paid a $10,000 fine.)