One power event at a single Google data center in London locked millions of people out of their bank accounts, stalled court hearings, and shut down payment systems across Europe Tuesday morning — and that is exactly the fragility you get when essential infrastructure runs through a handful of tech giants.

The outage hit just after 6 a.m. GMT in Google's europe-west2 region, taking down services at Lloyds, Halifax, HSBC, Bank of Scotland, and the digital bank Revolut, according to WTOP and Protos. Revolut's own status page showed the full scope: ATM withdrawals, card payments, transfers, identity verification, sign-in, chat support, and even stocks and crypto trading — all down. The bank posted on X that "all our customers' money and data are safe," but when you can't withdraw a dollar from an ATM, that assurance rings hollow.

It wasn't just banks. The Guardian reported that the Cloud Video Platform used by courts across England and Wales went dark, delaying hearings and preventing defendants in custody from appearing remotely. Staff at Southwark Crown Court called it a national outage. Delivery service DPD and Royal Mail were also hit. ING Germany reported separate online banking problems, per CNN, though the connection to Google wasn't confirmed.

Google blamed a "power event" at its europe-west2-b zone. The company said the event was mitigated and services would recover gradually, according to The Verge. Google's own suggestion to impacted users? "Migrate their workload impacted to another available location (zone or region)," Protos reported. Move your whole operation, in other words — as if that's something a bank or a court system can do on the fly.

The legal reality here should alarm anyone paying attention. Ryan Gracey, a partner at law firm Gordons, told The Guardian that most cloud contracts "protect the supplier far more than they protect the customer." The typical remedy when a provider goes down is "a modest credit against future fees" — small consolation for lost revenue and damaged reputation. "Liability for those wider losses is typically capped or excluded altogether," Gracey said. The real exposure, he warned, sits in "the gap between how heavily a business depends on its provider and how little its contract actually guarantees."

This isn't a one-off. Protos noted that a Google Cloud outage last September disrupted AI companies, and an Amazon-related cloud outage in April 2025 forced major crypto exchanges to pause withdrawals. The pattern is clear: centralize the world's infrastructure in three cloud providers, and one blown transformer can freeze an economy.

The Guardian framed the story around business exposure and contract fine print. CNN and WTOP kept it to bank disruptions and Google's status updates. Protos rightly highlighted the cascading scope — crypto, trading, ATM access — and the prior outages. Nobody in the mainstream press asked the obvious question: why does a single corporate data center control whether a citizen can buy groceries or attend a court hearing?

Regulators expect banks to keep essential services running even when a key supplier fails, as Gracey noted. But "outsourcing a service does not mean outsourcing responsibility for it." Right now, the responsibility sits with companies that write liability caps into their contracts and tell customers to migrate workloads when the lights go out.

Google's engineers will fix the servers. The deeper outage is architectural — a system where one provider's failure is everyone's problem, and the customer is last in line for recourse.