U.S. Insurance Sector Loses 90,000 Jobs; AI Is Not Sole Cause

The employment report released by the Bureau of Labor Statistics (BLS) on Friday (2) provided an unsettling number for the insurance sector. Employment in financial activities in the United States is 129,000 jobs below the peak in May 2025, with approximately 90,000 of these losses occurring in insurance companies and related activities, according to the BLS. In September alone, the financial sector as a whole shrank by 7,000 positions, in a month when the U.S. economy added only 29,000 jobs.

Following consultancies and banks, insurance is the third white-collar sector experiencing the same tension: increased investment in AI alongside a declining workforce. However, the cause-and-effect relationship is less obvious than the numbers suggest.

Cuts Were Greater Than Planned

The Insurance Labor Market Study for the third quarter, conducted by the Jacobson Group and Aon, highlights the gap between intention and practice. In July 2025, 13% of property and casualty insurers indicated they planned staff reductions; a year later, 26% reported they had made cuts. Among life and health insurers, 10% planned cuts, but 29% executed them. Automation has become the most cited reason for those still intending to cut jobs.

On the European side, Allianz Partners, Allianz's travel insurance and assistance branch, confirmed in July that it will eliminate between 1,500 and 1,800 positions in Europe in 12 to 18 months, mainly in call centers. According to the CEO of the unit, Tomas Kunzmann, the increasing use of AI is driving this reduction. The unit has approximately 22,600 employees, of which about 14,000 handle customer and claims inquiries over the phone. In the U.S., broker Acrisure announced in May that it would cut 2,250 jobs, or about 11% of its global workforce, citing advancements in AI and automation.

The Evident AI Index for Insurance showed in June that AI-specialized roles grew by 32% in one year, now accounting for nearly one in every 50 employees in the 30 evaluated insurers, while the total workforce shrank by 2.2%. The German company ERGO, part of Munich Re, created the position of Chief AI Officer on its management board, which has been held since October 1st by Guy Goldstein.

The Counter Argument Deserves Consideration

Robert Hartwig, a finance professor at the University of South Carolina and former chief economist at the Insurance Information Institute, asserts that the direct link between AI and job losses is more nebulous than it appears. He recalls that in 1999, when insurers launched online sales portals, the market also predicted the extinction of brokers, which did not occur. Part of the decline can be attributed to high-interest rates, and much of the contraction happens without mass layoffs: positions that are not replaced and fewer entry-level roles being opened with each cycle.

Other data weaken the substitution thesis. In the same study by Jacobson and Aon, 49% of insurers plan to increase their workforce in the next 12 months, compared to 11% who intend to reduce it. An industry being hollowed out by AI would not have more than four times as many companies wanting to hire as cut jobs. The two symmetrical errors are apparent: treating each lost position as proof of replacement and considering the 1999 historical precedent as a guarantee that nothing will change now.

The distinction often overlooked in the debate is where the loss is concentrated. Replacement occurs in high-volume, low-variation roles, like phone support and claims triage. Hiring is happening in engineering, data, and model governance. The net negative balance is real, but the crucial question for a board is not whether AI is cutting jobs in insurance. It’s which roles the company will stop replacing and whether there are people to operate what comes in their place.

Where the Effect Reaches: India and Europe

The impact is not confined to the U.S. A large part of the back-office operations for American and European insurers is run from operation centers in India, outsourced to companies like WNS, based in Mumbai and specializing in insurance processes. Capgemini acquired WNS for $3.3 billion in 2025, employing over 64,000 staff, presenting the acquisition as a foundation for "intelligent operations" powered by agile AI. When software agents take over customer support and claims triage, the first positions affected tend to be the offshore analysts, not those at headquarters.

In Germany, Allianz’s situation illustrates the other channel: European call centers that, according to the company, are adjusting through severance agreements and early retirements instead of direct layoffs. The pace is slower; the direction, the same.

The shared risk among the three markets is the career ladder. Insurers that cease to open entry-level positions in claims and support save money now and discover, in five years, that they have not trained the senior underwriters and claims adjusters that AI has yet to replace.