CHAUDHRI: AI use leads to legal trouble for this major tech employer
· Toronto Sun
While I usually write about employees’ use of AI in the workplace, this week I have set my sights on a recent lawsuit brought by 26 Meta employees in northern California.
The group claims that they were all laid off from their jobs at Meta after the company used AI to decide which workers should be let go.
Visit milkshakeslot.com for more information.
The claim, according to the Wall Street Journal , is that AI disproportionately targeted workers with disabilities or those who were on medical or parental leave.
One plaintiff who was laid off, claimed she was on a pregnancy leave and was informed of her layoff two days before her delivery. At least one other Meta employee was laid off while on a medical leave for a “serious health condition.”
According to the WSJ article covering the lawsuit, the group of workers claimed Meta’s AI tools “draw on inputs-performance ratings, calibration scores, productivity and output metrics, ‘AI-native’ ratings, and AI-token consumption-that, by design, cannot be accumulated by an employee who is on protected medical or family leave, or whose output is reduced by a disability.”
The argument is compelling. When employers use AI tools to track performance and identify “under performers,” employees who are absent from the workplace on protected leaves or are working in a limited capacity due to an accommodation or medical issue could rise to the top of the list in disproportionate numbers. That could be a discriminatory practice and a dangerous one at that, if gone unchecked.
If you, the employee need to take a month off of work to treat an illness, only to be flagged by your manager as an under performer for that quarter, you would likely point out the inherent unfairness in that practice.
Life, after all, can be unpredictable and employers should not be permitted to seize upon a job-protected absence to give an employee a failing performance grade.
In my own practice, I regularly work with executives who are subject to annual targets or quotas. If any of these individuals had to take a medical or parental leave (and some have), I noticed that in many cases their annual targets were not adjusted to reflect their time away. Instead, they were expected by their employers to “make up” for the absence after their return by working more hours, still expected to reach their annual targets.
Naturally, this is rarely possible. If an employee is away from the workplace for 12 weeks, could they reasonably reach their annual targets working only three quarters of the year? Probably not.
RECOMMENDED VIDEO
Absences are disruptive and unpredictable
Why would employers do this? Absences are disruptive and unpredictable. Particularly in high growth industries like tech. By rating employees who take protected leaves as “under performing” employers change the narrative, and shift blame in order to avoid discriminatory claims if the employee eventually is laid off or terminated.
While many tech companies use sophisticated tools to measure performance, in practice, most employees feel disempowered around performance reviews and often believe they cannot challenge low ratings or performance scores. The reasons given to support the low rating can be difficult to understand and navigate (especially if an AI tool has been used).
In fact, employees should never accept a lowered performance rating if objectively the rating does not match their performance for that period. If an employee is concerned that their negative rating is tied to an absence, they should consider challenging the rating and asking for the back up to support the decision.
When employees don’t challenge these reviews they could be seen to accept their rating and this can be much more difficult to overturn if that same employee is terminated soon thereafter.
While employers may use AI to manage employees, they still own the liability that flows from the decision-making. People, after all, are still needed to manage people.
Have a workplace problem? Maybe I can help! Email me at [email protected] and your question may be featured on a future column.
The content of this article is general information only and is not legal advice.