Apollo's Marc Rowan is right that insurance capital will power the AI boom. He's wrong if he thinks capital and appetite are the same as proper risk governance — and UK regulators will punish firms that confuse the two.
Industry View  Trovix AriaInsurance · Financial Services

Marc Rowan's assertion that insurance and retirement capital could finance an $8 trillion AI buildout is significant for UK regulated firms because it confirms what mid-market insurers, law firms and financial services businesses already sense: AI infrastructure financing is now a core asset class, and the firms with capital are the ones deciding who gets access. For UK-regulated insurers especially, this represents both opportunity and a regulatory minefield. The FCA's Consumer Duty (PS22/9) and the PRA's Supervisory Statement on Third-Party Risk (SS1/23) have already made clear that capital decisions must be paired with rigorous governance. Betting capital on AI infrastructure without understanding the underlying deployment risk — how the AI is actually being built, by whom, and what it can fail at — is exactly the kind of hidden tail risk that regulators will examine in supervision and that will show up in a breach.

What Rowan's comment reveals is that the financial services industry has stopped asking 'is AI ready?' and started asking 'who has capital to deploy it?' That is a dangerous inversion. We are seeing it everywhere: Harvey and Legora raising capital and deploying to law firms without solving hallucination in citation; Luminance and others packaging data extraction as AI governance when it is just pattern matching; enterprise Copilot deployments failing silently because nobody integrated them with actual compliance workflows. The pattern is consistent: capital flows, tools proliferate, and 18 months later someone has to explain to the FRC (through ISA UK audits) or the ICO (through UK GDPR breach reports) why the AI tool produced wrong advice or exposed client data. Insurance financing makes this cycle faster and bigger. It does not make it safer.

Trovix's view is straightforward: capital for AI infrastructure should never move faster than governance for AI use. That is not conservative; it is arithmetic. If an insurer or law firm is taking capital to deploy AI and that capital is not explicitly paired with AI risk assessment, compliance logging, and outcome monitoring — the kind of governance that Trovix Audit is designed to provide — then the firm is taking on undisclosed regulatory risk. The EU AI Act, ISO 42001 and the emerging UK AI Bill will all require documented impact assessment and audit trails. Firms that have deployed AI without that framework now have to retrofit it. Firms that accept capital without it built in from day one will have no defensible record when a regulator asks 'what did you know, and when did you know it about that model's performance?'

For a mid-market law firm, accountancy practice or insurance broker right now, the practical step is not to chase insurance-backed AI financing. It is to audit what AI systems are already in use and codify the risk assessment and performance monitoring around them before capital decisions accelerate the problem. If your firm is using generative AI for document review, client communication, tax research or underwriting — Trovix Sift and Trovix Aria let you inventory what is running and log how it is performing. If capital is on the table, insist that AI governance funding is part of the same conversation. The firms that will survive regulatory scrutiny in this cycle will be the ones that made the boring, expensive choice to govern AI properly before they had someone else's capital to spend.

Source: Bloomberg News

Related Trovix product:

Trovix Aria →Book a demo →