OpenAI's new Codex plugins for finance and legal work sound impressive until you ask the question that actually matters: who is liable when an AI agent makes a decision in a regulated firm? The answer is never the AI vendor. Yet most firms still buy AI tools as though it is.
Read more: OpenAI's Sector Tools Won't Fix Regulated Firms' Real Problem
Morgan Stanley is opening its wealth platform to thousands of external AI agents—a bet that interoperability outweighs governance risk. UK firms cannot wait for regulation to catch up; they must decide now whether to follow or build defences first.
Read more: Morgan Stanley's AI Agent Play Exposes UK Firms' Compliance Gap
The UK government is embedding unaudited AI into legislation while spending half a billion on consultancy with no visible controls. Regulated firms must build governance frameworks the state has not yet mastered.
Read more: Government AI risks prove private sector must lead
UK firms are deploying agentic AI at scale without the governance frameworks that regulators now demand. This is not a technology problem. It is a compliance crisis that will hit hardest those firms still pretending governance can wait.
Read more: 87% deploy AI. 75% have no governance. That's a crisis.
OpenAI's vertical expansion into legal and finance is technically impressive and strategically irrelevant to the real problem facing UK regulated firms. Governance matters more than vendors.
Read more: OpenAI's Vertical Pivot Won't Solve Your Real AI Problem
Freshfields and Quinn Emanuel are running live client work on Claude. They should not be. Not yet. The gap between what these tools can do and what firms have put in place to verify they actually do it is now a regulatory time bomb.
Visa's new dispute AI tools reveal a dangerous trend: automating regulated processes without building in governance. Mid-market UK firms need to stop chasing speed and start demanding explainability.
More Articles …
Page 52 of 72