The Cost of Waiting Falls Hardest on Women. Here’s How We Change That.
Three reports published in the past six months tell this story: AI is changing what most American jobs look like, faster than employers are adjusting. And the disruption is unevenly distributed.
The Headline Number. Athena Corporate Partner, Boston Consulting Group, in an analysis of roughly 165m US jobs released earlier this year, projects that AI will substantially reshape 50-55% of American roles within two to three years. It could eliminate a further 10-15%, or 16m-25m positions, within five. “Reshaped” is the operative word here. Most keep the title as the day’s work changes underneath them.
The HR View. SHRM’s State of AI in HR 2026, surveying 1,722 HR professionals, sharpens what this “reshaping” feels like inside companies:
39% of organizations have deployed AI in HR; another 7% plan to this year.
57% of HR leaders report frequent AI-driven upskilling, and 39% have seen job duties shift.
Yet only 49% have a formal AI policy, and 25% call theirs future-proof. HR staff in states with AI employment laws on the books, 57% claim they do not know those laws exist.
Adoption is Running Ahead of Governance. Training is running behind both.
The gender lens. The World Economic Forum’s March update on AI and gender parity reports that women hold 57% of the US roles most exposed to generative AI, against 43% for men. The International Labour Organization puts it sharper: 29% of jobs in female-dominated occupations are exposed to generative AI, against 16% in male-dominated ones. Of the 6m US workers least likely to recover from an AI-related job loss, 86% are women. Tech itself is not safe ground. Women's share of European tech jobs fell from 22% in 2023 to 19% this year.
Why This Matters Here. Athena members sit at the intersection of those trends. As the work is being reshaped, the rules have not caught up, and the cost of inaction falls hardest where representation is already thin.
A CASE STUDY:
BANK OF AMERICA
The bank treated AI as a redesign question, not a benefits one. It built an internal Academy that reskills staff across the AI stack, from basic prompt engineering to design and development.
Illustration: Alberto Miranda
The result: more than 14,000 employees moved into new roles inside the company in 2025, and 44% of all positions over the past few years were filled internally rather than from outside. Its employee AI assistant, Erica for Employees, now reaches more than 90% of staff and has cut calls to the IT service desk by half. The lesson is operational, not aspirational: when reskilling is plumbed into mobility, retention and productivity move together.
Three things to do this quarter…
Audit AI exposure role by role, not department by department. BCG's data shows reshape rates vary within the same function.
Make upskilling a default, not an opt-in. SHRM finds training is happening; it is just fragmented.
Write the AI policy your state already requires. Most HR teams in covered states do not yet know they are covered.
The firms that advance over the next 24 months will treat this as a redesign opportunity, not a training problem.
Athena AI Academy
JOIN ATHENA AI ACADEMY
LAUNCHING ON JULY 28
Twenty seats cap. Built for leaders who aren't waiting to be reskilled by someone else. Learning in community. Intimate cohort size. A capstone project and the ROI to take back to your company — and your board.
For those who refuse to give away your voice or your agency as AI and other market forces bring waves of uncertainty, kindly complete this 4 question survey HERE and shape Athena your way.
Until next time,
