Capital and regulation are converging on agentic AI in finance. Mega-rounds like Ramp’s $60bn valuation show investors still pouring money into AI-driven fintech, while London’s insurance market is moving to narrow cyber cover until it can price agentic AI risk. UK regulators, meanwhile, are testing AI agents in live markets rather than writing AI-specific rules.
Top story: Fintech Ramp reportedly hit a roughly $60 billion valuation in a $1.85 billion round led by Dragoneer and Thrive Capital.
Ramp hits $60bn valuation in $1.85bn AI-fintech raise
Bloomberg · Finance
Corporate-spend fintech Ramp has reportedly reached a valuation of around $60 billion, raising about $1.85 billion in a round led by Dragoneer and Thrive Capital, with Founders Fund also contributing. The scale of the round shows investor appetite for AI-powered finance automation remains intense, and it sets a new valuation benchmark for competitors and incumbent banks’ business-spend products.
London cyber insurers weigh limiting cover over agentic AI risk
Resultsense (citing City A.M.) · Risk
Some London cyber underwriters are considering restricting cover until they can measure and price agentic AI risk, and the Lloyd’s Market Association is drafting a standard AI definition. For UK businesses the practical consequence is that AI-related exclusions or sub-limits could appear at renewal, so policy wording needs close checking. Note this report is roughly two weeks old, the freshest Lloyd’s item surfaced.
https://www.resultsense.com/news/2026-09-24-lloyds-market-cyber-insurers-agentic-ai/
Bank of England to test AI agents ‘herding’ in markets
UK Parliament Treasury Committee · Regulation
The Bank of England has confirmed plans to test the use of AI agents in trading markets, investigating whether they could behave in correlated ways, known as ‘herding’. HM Treasury has not committed to bringing major AI and cloud providers into the Critical Third Parties regime before end-2026, despite MPs’ criticism. This is the UK’s clearest signal that AI-driven market instability is now a supervisory priority (published April 2026; the most recent official UK position found).
FCA puts Barclays, UBS and Lloyds AI under live testing
FCA · Tools
The FCA’s second AI Live Testing cohort includes Barclays, Experian, Lloyds Banking Group (Scottish Widows) and UBS, trialling use cases from agentic payments to anti-money laundering detection and KYC. Testing runs to year-end with a good and poor practice report due later in 2026, and the FCA is relying on existing rules rather than bespoke AI regulation. Practitioners should expect that report to shape supervisory expectations.
https://www.fca.org.uk/news/press-releases/fca-announces-second-cohort-ai-live-testing
Anthropic launches ten AI agents for banks, insurers and PE
Forbes · Generative AI
Anthropic released 10 agent templates for pitchbooks, credit memos, KYC screening and month-end close, alongside a $1.5bn joint venture with Blackstone, Hellman & Friedman and Goldman Sachs to embed Claude in portfolio companies. It signals AI vendors moving from selling to banks to operating inside them, with direct implications for junior analyst workloads and private equity value creation (May 2026 launch; the latest major finance-specific vendor move surfaced).