Signals · Labour Market
US insurers are 95,000 jobs below their peak, and AI savings across companies have not caught up
US insurers have cut 95,000 jobs since February 2025 and more firms cut than planned, while Bain finds company AI savings trail their targets.
by Jo·4 min read·
New here? Start with the free AI Survival Kit →
US insurers are 95,000 jobs below their peak, and AI savings across companies have not caught up
0:00 / 5:59
This voice is generated by AI.
Part of the guide Will AI take your job? How to find out for your role, in an afternoon
Insurance Business reported on Monday that the US insurance industry has now cut jobs for 11 consecutive months. US Bureau of Labor Statistics figures released on Friday show that carriers and related businesses shed another 2,300 jobs in September. That leaves about 2.93 million workers, roughly 95,000 fewer than at the February 2025 peak. The industry has now lost more jobs than it did between July 2008 and early 2011. In percentage terms the fall is smaller, 3.2% against 3.6%, because the workforce is larger now. The piece, by Matthew Sellers, asks whether Australia should worry. A more useful question is what insurers are actually cutting, and whether the cuts are paying off.
The plans ran ahead of themselves
The sharpest number in the report comes from the Q3 2026 Insurance Labor Market Study by The Jacobson Group and Aon. In July 2025, 13% of US property and casualty respondents said they planned to reduce staff. By the following July, 26% said they had done so. Twice as many firms made cuts as had said they would.
Among companies still planning cuts, automation is now the most common reason. It was cited by 17% of all participants, ahead of overstaffing on 14% and reorganisation on 12%. Some firms say so outright. Acrisure pointed to AI and automation when it announced 2,250 job cuts in May. Allianz's chief executive confirmed cuts that the company puts down to AI. This followed earlier reports that its travel insurance arm, Allianz Partners, planned to cut up to 1,800 roles worldwide.
Other firms give different reasons. Acenda is cutting at least 280 roles as it combines the former MLC Life and Resolution Life businesses, and middle management is bearing the brunt. In the UK, Admiral said its plan to cut about 500 roles was about efficiency rather than AI. The BLS data does not record why anyone lost a job, so the 95,000 cannot be pinned on a single cause.
The usual reading takes that uncertainty as reassurance. It should be read the other way. A merger, an efficiency drive and an automation programme all ask the same question of a role: which parts of it still need a person? Middle managers at Acenda faced that question whatever label went on the announcement.
The savings have not arrived
Insurance Business cites two pieces of research that should trouble every executive running these programmes. Gartner found that firms getting strong returns from autonomous technology were cutting jobs at almost the same rate as firms getting modest or negative returns. Bain & Company's 2026 Automation and AI Pathfinder Survey found that nearly 40% of companies measuring their AI cost savings had saved less than 10%. Their targets were between 11% and 20%.
Taken together, the two findings say headcount is falling whatever the results. Firms are making the cuts before the systems have shown they can do the work.
That gap is where an operator can strengthen their position. When a firm lets people go before the tools deliver, someone still has to handle the judgement calls, the exceptions and the client relationships the system was meant to take over. AI does the routine work. You do the thinking. The person who can show the work still gets done that way is the one the next restructuring keeps.
Your Next Move
- Map your role before the planning cycle does. List a week of your work. Sort it into tasks a system could run on your instructions and tasks that need your judgement. Our step-by-step method for finding out whether AI will take your job takes an afternoon.
- Measure what the tools actually save in your team. According to Bain, nearly 40% of firms that measure their savings got less than 10%. Bring a real figure for time saved and errors caught to your next review. Whoever holds the evidence shapes the decision.
- Own the exceptions. Volunteer for the cases automation gets wrong: complex claims, disputed decisions, the client who needs a phone call. That work stays when the routine work is automated, and the people who decide the next round can see who does it.
I build systems like the one publishing this site. → Work with me
About the author
Jo
Jo runs The War Room: one signal a day on how AI is changing work, and what to do about it.
Sources
Get the Briefing
Want more intelligence like this?
The weekly briefing, free — and the AI Survival Kit with it.
More in Signals
See allSignals · Capability Displacement
Utah's AI prescribing pilot shows how a profession hands over a task in three phases
3 min read
Signals · Governance
OpenAI's safety lead quit over culture, and his fix is the human oversight every team running AI agents now needs
3 min read
Signals · AI Adoption
A two-year Khanmigo trial shows that access to AI is not the same as using it
3 min read