An insurance broker being bought for $7.7 billion on an AI bet says AI helped with 47,000 tasks. The savings it has shown are small so far

Trevor Baldwin, chief executive of insurance broker The Baldwin Group, used a Fortune commentary piece on Nov. 13 to argue that AI will remake the insurance advisor's job. His main evidence: more than 47,000 AI-assisted tasks in 17 weeks, at a quality rate above 98%.

The stakes go well beyond one essay. In September, Tampa-based Baldwin agreed to be taken private by Sequence Holdings and DFO Management, Michael Dell's family investment office, in a deal valued at about $7.7 billion. Shareholders get $32.50 a share in cash, an 88% premium to the price before deal reports surfaced.

The buyers pitched the deal as a way to rebuild the firm around AI, with long-term capital and no quarterly earnings pressure. That makes Baldwin one of the largest public test cases of an AI-led overhaul. Its disclosures so far show real gains in narrow back-office processes, and little yet at the scale of the price tag.

What the numbers actually show

The 47,000 figure is not new. Baldwin first gave it on the company's July 30 earnings call, for the same 17-week window and the same 27 standardized processes. On that call he said quality had climbed above 99% in recent weeks. The Fortune piece repeats the older 98% figure and adds no newer totals.

The most concrete example is direct-bill reconciliation: matching commissions that insurers pay directly against what the broker expects to receive. Baldwin said the share reconciled within the month rose from about 90% to a sustained 98%. The process's annual running cost fell from $3 million to $1 million, and internal labor cost from $1.2 million to about $400,000.

The Fortune piece adds one new anecdote. An early claims-review tool spotted coverage gaps in minutes, which led to $71,000 in additional claim payment for a client. Baldwin notes that a professional still reviewed the findings. None of the company's public statements reviewed here defines a "task" or a "quality rate," and none of the figures has been independently checked.

Scale matters. Baldwin booked $1.5 billion of revenue in the first nine months of 2026. A $2 million annual saving on one process is useful, but small against that base. A year ago, Baldwin told investors its transformation program targets $40 million in annualized run-rate savings by the end of 2028, against roughly $40 million in one-time transformation charges.

What the $7.7 billion is riding on

Baldwin spent $12.4 million on transformation costs in the first nine months of 2026, including $2.9 million in the third quarter, according to its Nov. 9 results. Those costs, plus $21 million of third-quarter deal expenses, are excluded from adjusted earnings. On a standard accounting basis, the quarterly net loss widened to $81.6 million from $30.2 million a year earlier.

The operating picture is mixed. Third-quarter organic growth, which strips out acquisitions, was 6%, up from 2% in the second quarter. Adjusted profit margin before interest, taxes, depreciation and amortization rose to 22.2% from 19.8%. But nine-month organic growth of 3% was the slowest since Baldwin listed in 2019, Insurance Business reported. Most of the 31% jump in total revenue came from buying CAC Group.

In July, Baldwin said AI gains were "very much embedded in the numbers." He did not put a number on AI's contribution to margins or earnings, and said the company was "still very early days."

Sequence's broader thesis is that the surest way to capture AI's value in some industries is to buy strong incumbents and rebuild them, rather than wait for startups to displace them. Co-founder Michael J. Lee has cited BankSouth, a Georgia community bank Sequence invested in, where he says consumer loan underwriting time fell 94%. Those figures are also self-reported.

What buyers and investors should watch

Baldwin's essay frames the payoff as capacity rather than cost cuts: advisors freed from paperwork to spend more time on judgment and client strategy. That is a common pitch. It is also harder to audit than a savings line.

Peers offer a benchmark. Rival broker HUB International rolled out Anthropic's Claude to more than 20,000 employees and reported an average of 2.5 hours saved per employee each week. Baldwin, which expanded its own firm-wide Claude deployment in May, has not published a comparable per-employee figure.

Three signals will show whether the AI case holds. First, whether the task count keeps growing and gets clearly defined. Second, whether organic growth and margins improve without acquisitions doing the work. Third, whether the $40 million savings target is raised before the company goes private and public reporting largely stops. The deal is expected to close in the first quarter of 2027, pending shareholder and regulatory approval.

For executives weighing a similar program, Baldwin offers a realistic baseline so far: solid gains in repetitive matching work, a sizable one-time bill, and a long wait before the benefits show up at company scale.