Adviser shortage: the generation of 2040 is already born
29.7% fewer passed Wft exams in five years, a rapidly ageing advisory profession and smaller birth cohorts. What that means for capacity in 2040.
Blog
Practice & Innovation
Field reports on AI experiments, innovation, insurance, search technology and day-to-day execution.
Insurance software 2026: the broker becomes the supplier
Two brokers are building AI software their peers have waited for. Why insurance software competition now comes from vendors' own customers.
Eleven years later: the diploma shortage I saw coming, and the blockchain promise that never arrived
In 2015 I said the profession had become less accessible because of diploma requirements. Eleven years later, CDFD data backs that up while my blockchain prediction missed completely.
Professional indemnity insurance and AI: zero hits
Is AI damage covered by your professional indemnity insurance? I searched 9 Dutch policies: AI is mentioned nowhere. What that means and what to arrange now.
The future of the insurance broker: your license is your right to exist
AI will soon advise better than you, but brokering requires a license. Duty of care stays with you — that's the broker's right to exist.
AI Overview Liability: Is Your Search Traffic Coming Back?
A German court held Google liable for false AI Overviews. I'm measuring whether that brings search traffic back on medical and legal queries.
What OpenAI and Google announced this week won't stop insurance fraud. What will?
OpenAI joined C2PA and SynthID on May 19. My tests with two fake receipts show that neither OpenAI's verification tool nor Gemini detect the SynthID watermark from Google's own Gemini Nano Banana 2. What fraud departments should build instead.
ChatGPT refuses fake receipts. Codex (same company) makes them anyway.
A year after my fake-receipt test, I ran the same prompt past ChatGPT, Codex and Gemini again. One company, two faces, and a new kind of AI safety you couldn't see a year ago.
Autoresearch: the end of A/B-testing as we know it
On March 6, 2026, Andrej Karpathy published a 630-line Python script. Within a week it had 30,000 stars. It's called autoresearch — and it changes everything we thought we knew about optimisation.