Microsoft and Chase Outages May Trace to AI Patch 'Mad Rush'
A wave of high-profile outages at Microsoft, JPMorgan Chase, and Charles Schwab has officials pointing not at a cyberattack but at companies patching AI-found flaws faster than their systems can absorb.

Kevin Mandia, CEO of cybersecurity firm Armadin, and Sen. Mark Warner say the recent Microsoft, JPMorgan Chase and Charles Schwab outages may stem from a faster-than-normal push to patch AI-found software vulnerabilities, not a coordinated cyberattack. For executives, the takeaway is that AI-driven remediation can create operational risk, forcing boards to weigh patch speed against testing capacity and system stability.
The outages that hit Microsoft 365, JPMorgan Chase and Charles Schwab over the last five days may not be the work of a coordinated cyberattack. They may be collateral damage from a faster-than-normal push to patch software vulnerabilities discovered by AI, according to politicians and tech executives who have looked at the pattern. Kevin Mandia, CEO of cybersecurity company Armadin, put the trade-off bluntly on CNBC's "Squawk Box" on Wednesday: "We're updating to withstand and be more resilient, and sometimes you do that at a speed where you couldn't test all the potential outcomes of the update."
Sen. Mark Warner, vice chairman of the Senate Intelligence Committee, added the political weight. He reiterated assurances from other government officials that the outages were not the result of a coordinated cyberattack by another country or malicious actor. But he told CNBC on Wednesday that AI tools can identify vulnerabilities in existing tech stacks, prompting a "mad rush" to create patches. "When that happens, careless things happen, the system can't absorb those patches as quickly, and you have these series of outages that have taken place," Warner said. "I think we will see more."
The past week has been a stress test for digital infrastructure. In five days, outages at Microsoft, JPMorgan Chase and Charles Schwab caused major disruption for businesses and consumers. A temporary ground stop at every airport in America also caused mass chaos, though Department of Transportation Secretary Sean Duffy blamed aging air traffic control infrastructure for that issue. Tech outages have long occurred periodically. What changed is the proximity: the cluster of incidents led many to speculate about a coordinated cyberattack, rogue AI, or a combination of the two.
The less cinematic explanation is more unsettling for operators. AI can discover cyber vulnerabilities that have remained undetected for years in human-made, occasionally sloppy software. There are also concerns that the technology could be used to exploit security flaws more quickly than humans ever could. Both fears appear to have prompted companies to patch more than usual, which in some cases may have caused things to break. The stakes rose last month after AI discovered huge flaws in a programming language that up to 70% of websites use. Since then, there have been a crypto heist, refrigerator hacks and attempts to literally boil blood. That is the kind of run of incidents that makes security teams want to move fast.
For peers in similar roles, the pattern matters beyond the immediate headlines. Public companies typically run change advisory boards, staged rollouts and maintenance windows for a reason: complex systems fail in coupled ways. AI can compress the time between discovery and deployment, but it does not automatically expand testing capacity, on-call staff or rollback options. That gap is where outages live.
Mandia expects the pace to continue. Organizations will likely keep updating their software at a faster rate, he said, and that will probably lead to more tech outages. "Nobody wants an outage," he said. "Everybody's going to get operationally better at updating in the AI age. We're just at the first inning of this." He argued the near-term cybersecurity pain will likely result in long-term gains.
Nicole Perlroth, a former cybersecurity reporter and founder of a cybersecurity-focused venture capital fund, made a similar point in an X post. She said new pressure to patch issues and update software faster means those fixes are more likely to "break things in unexpected, cascading ways." That is the second-order risk boards need to price: security teams are being asked to reduce breach exposure with AI speed, while operations teams are still measured on uptime. If the two goals are not governed together, the fix can become the incident.
For executives, the lesson is not to slow down security work. It is to treat patch velocity as an operational capability that needs testing, staging and rollback plans, especially at companies whose systems customers touch every day. A bank app, a productivity suite or an airline operations center cannot absorb an unlimited rate of change. The companies that win the AI security era will be the ones that can patch faster than attackers without breaking the services their customers rely on. Warner's warning that more outages are coming is a reminder that resilience is now a board-level metric, not just an IT ticket.
This story's Key Insights and Take-aways are locked.
Create a free account to unlock Executive Actions for one credit.
Register to UnlockAlways free for Executives Club members. Join the Club
More in Technology
Anthropic mathematician claims Claude helped crack the Riemann hypothesis
If the proof holds, AI has toppled math's most famous open problem; if it does not, it is a case study in announcing breakthroughs before anyone verifies them.
AI's $4 Trillion Wipeout Erases Nasdaq's Gains for the Year
A brutal week took roughly $4 trillion off the AI sector and left the Nasdaq flat for the year, and the damage ran the full length of the supply chain.
Anthropic mathematician says AI proved the Riemann hypothesis. Verification will take time.
A single unverified claim from inside Anthropic now sits against more than 160 years of open mathematics, and the gap between asserting a proof and surviving peer review is the entire story.


