Summary: The Age Discrimination Machine
How Algorithms, Culture, and Law Fail Older Workers
She applied for 143 jobs over eleven months. MBA, twenty-eight years of brand marketing experience, revenue growth at three companies. Seven interviews. All stalled after the video call. One interviewer asked when she graduated. Another asked if she would be comfortable reporting to someone younger. The rejections cited culture fit, or said nothing at all. She is fifty-seven.
Age discrimination does not arrive with a note explaining itself. Resume screening, increasingly automated, filters applications before any person evaluates them. Job postings seek “digital native” skills and “startup mentality.” “Overqualified” means too old, expressed in legal language. The system leaves no fingerprints: when you apply and hear nothing, you cannot prove what happened.
Algorithms compound the problem. Applicant tracking systems trained on historical hiring data learn that youth predicts success, because companies historically hired younger workers. The algorithm screens out proxies for age: graduation years, decades of experience, employment gaps. Meta settled with the DOJ after its tools allowed employers to target job postings by age. Most algorithmic discrimination is invisible, embedded in proprietary systems never audited.
The Age Discrimination in Employment Act protects workers 40 and older, but the Supreme Court’s 2009 Gross decision raised the burden of proof: plaintiffs must show age was the “but-for” cause, a higher standard than applies to race or sex claims under Title VII. Proving discrimination that leaves no trace is nearly impossible. The EEOC receives 12,000 to 15,000 age charges annually; most result in no-cause findings. For the marketing director screened out of 143 jobs, there is no case to bring. She has frustration, not evidence.
The Protecting Older Workers Against Discrimination Act would restore the mixed-motive standard. It has been introduced repeatedly. It has never passed. Algorithmic accountability is emerging in fragments: Illinois requires notification when AI is used in video analysis, New York City mandates bias audits for automated hiring tools. Federal action is sparse.
The gap between what research shows and what employers do is not a gap of knowledge. It is a gap of incentive, culture, and power.