JPMorgan Chase said transactions were posting normally on Monday evening after a software bug in its own hardware systems disrupted online banking, branch transactions and Zelle payments across the U.S.

Chase said on X at about 4 p.m. Eastern time that it had "experienced a temporary issue related to a software bug in on-premise hardware systems". A spokesperson said there was no evidence the problem was cyber-related and that it was not systemic. Chase said customers kept access to their accounts and funds throughout.

An update to Chase's own physical computer systems inadvertently caused the fault, a person familiar with the matter said. Customers also reported trouble with transactions at Chase branches, where some staff told them the bank's internal system was failing. The disruption began on Monday morning. A Chase spokesperson said at about 6:30 p.m. that transactions were posting normally again. Regulators contacted the affected banks immediately and quickly concluded that the problem was contained and that service was already being restored, according to a person familiar with the situation.

Downdetector, a website that collects user reports of service failures, logged nearly 4,800 reports involving Chase by 2 p.m., more than for any other bank. Some customers who opened the Chase app saw a notice saying some services were unavailable and asking them to try again later.

Reports also rose for Bank of America, U.S. Bank, Wells Fargo, PNC, Citigroup and Capital One, before beginning to decline by Monday afternoon. A U.S. Bank spokesperson said its systems were not directly affected but that clients sending or receiving Zelle payments with another bank could face delays.

Zelle lets customers of participating banks send money to one another within minutes. It is run by Early Warning Services, an Arizona company owned by seven large banks including JPMorgan Chase, Bank of America and Wells Fargo. An Early Warning Services spokesperson said the Zelle network was working normally and that a problem at the banks themselves may be delaying payments for some of their customers.

Microsoft and the FAA

Microsoft's Outlook email, Teams and Copilot services failed for users worldwide from about noon Eastern time on Thursday 12 November, with disruption continuing through the U.S. business morning on Friday. Microsoft attributed the failure to a problem during an earlier network capacity upgrade and said on Saturday that service had been restored. The company said the outage was not related to a cyberattack.

The Federal Aviation Administration halted departures across the U.S. at about 3:30 p.m. Eastern time on Friday 13 November. Its older equipment became unstable when data flooded its telecommunications network at a rate of 1,000 connection attempts every four to five seconds, disrupting radar feeds and communications with aircraft. The agency lifted the ground stop at 8:15 p.m. L3Harris, the contractor Transportation Secretary Sean Duffy blamed, said it was cooperating with the investigation.

The FAA had grounded aircraft nationwide only once before since the attacks of 11 September 2001. Duffy said L3Harris had been carrying out routine maintenance at a remote FAA site, switching telecommunications lines from copper wire to fibre optics. He said the FAA still relied on copper lines dating from the 1980s.

FAA Administrator Bryan Bedford said investigators had found "zero evidence of anything nefarious". More than 14,000 flights were cancelled on Friday and Saturday.

Trump was asked by a reporter outside the White House on Monday morning, before leaving for a summit in the Philippines, what responsibility he took for the ground stop given that he was president. "I don't take any responsibility," he said. He praised Duffy's handling of the ground stop, blamed old equipment and previous administrations and said he was working with Congress to get money for a replacement.

Outages since November 2025

Cloudflare, which carries and protects traffic for a large share of the world's websites, took X, ChatGPT and other services offline on 18 November 2025. Downdetector, which uses Cloudflare, also went down for a time.

Verizon's wireless network failed across the U.S. on 14 January 2026, disrupting calls and data for more than 10 hours. Emergency alerts in New York City and Washington warned that Verizon customers were having trouble calling 911. Verizon apologised and offered account credits.

Microsoft's Outlook, Teams and Defender services went down across North America on 22 January 2026 after part of the company's infrastructure in the region stopped processing traffic correctly. Most services had returned by early the next day.

American Airlines asked the FAA for a nationwide ground stop covering its own flights on the evening of 28 July 2026 after its internal systems failed. The stop lasted less than an hour.

Newark, Philadelphia, Teterboro, JFK and LaGuardia airports were placed under FAA ground stops or delays on 21 September 2026 after the primary communications circuit at the Philadelphia approach control centre failed. Its backup fibre line had been cut at a construction site in New Jersey. Duffy blamed an Amtrak crew, but Amtrak pointed to a New Jersey Transit crew working in the area. The same Philadelphia facility had disrupted flights at Newark in the spring of 2025.

Thanksgiving falls on 26 November. The FAA will handle the holiday travel peak on the telecommunications network that Duffy said went down everywhere at once on 13 November. Asking Congress for money to replace it, Duffy said: "This is an emergency." The Government Accountability Office reported in September 2024 that 51 of the FAA's 138 air traffic control systems were unsustainable.