In an increasingly interconnected world, some fear such disruptions could be the new normal. Technology failures reached email, workplace software, air travel, banking, consumer apps and the government network used for classified communications during the past week, according to accounts from the organizations involved. Problems at Microsoft, JPMorgan Chase and Apple, as well as the Federal Aviation Administration and National Security Agency, appeared to emerge from routine updates, security patches or other malfunctions. The Trump administration and the affected companies have each said they have no evidence linking the disruptions to a cyberattack.

Late last week, a global Microsoft 365 failure interrupted Outlook email, Teams and the Copilot AI assistant. Microsoft said the incident followed a failure during a recent upgrade to its servers and that the company later determined a “latent defect” in a network fix it had begun deploying over the summer was part of the cause. Microsoft did not specify the goal of that network fix.

A separate malfunction in the Federal Aviation Administration’s network produced the longest nationwide ground stop in 25 years on Friday. The stop lasted more than four hours, and carriers cancelled and delayed swaths of flights while they worked to recover. United Airlines then saw its own system outage on Monday, forcing departing flights to be delayed.

Online banking troubles piled up on Monday. JPMorgan Chase said a software bug affected its internal hardware systems, causing transactional delays for its customers and for others sending money to related accounts. The episode sent jitters throughout the financial system before the problem was resolved.

On Tuesday, Apple said an internal update temporarily disrupted Apple Music, the App Store and AppleCare services. Marriott said a technical issue caused problems for people signing in to its website and app. The National Security Agency said an internal network it uses to communicate classified information was offline Tuesday “during scheduled routine maintenance.”

The Trump administration said there was “zero” indication the latest outages were linked or caused by bad actors such as cybercriminals. The government and Microsoft, United, JPMorgan Chase, Apple and Marriott have maintained there is no evidence the disruptions are linked or related to a cyberattack. Microsoft also said its outage did not span from “AI / agents.” The Associated Press reported that Microsoft did not respond to a request for further details and that the other companies did not provide additional comments Wednesday when contacted by email.

Jay Clayton, President Donald Trump’s director of national intelligence, pointed directly to security patching when discussing the NSA outage. He told Fox News that patching following last month’s PHP alert caused some systems to “not come back up” as expected. “That is what has happened at the NSA today,” he said, while the NSA’s public account of the disruption described it as scheduled routine maintenance.

AI has also helped identify flaws. As AI tools uncover new vulnerabilities, companies and institutions have been racing to patch them before hackers can target them, a cycle that can lead to new glitches. AI recently helped detect sweeping vulnerabilities in PHP, a general-purpose scripting language that powers websites around the world, leading to urgent calls for updates, and that security work has resulted in some limited outages and service delays.

Also this week, cybersecurity startup Calif said it used AI to help uncover a software bug in Java, one of the world’s most widely used programming languages and crucial for many big companies. Oracle, which owns Java, later released a security update, and U.S. cyber authorities issued an advisory Wednesday urging organizations to install it.

Raj Rajarajan, director of the Institute for Cyber Security at City St George’s, University of London, said last week: “As AI adoption accelerates, we are likely to witness more incidents of this nature.” He noted that much of today’s digital infrastructure was not originally designed to handle the “computational demands of large-scale AI services.” The recent outages have not been tied to hackers or a coordinated attack, though Rajarajan and others pointed to broader cybersecurity risks.

Justin Cappos, a cybersecurity professor at New York University’s Tandon School of Engineering, put the risk more starkly. “AI is absolutely making it way easier for attackers to go and launch cyber attacks,” Cappos said. “It is not quite as good at creating patches for bugs, like fixing problems, as it is exploiting them.”

Google’s cybersecurity report for this week, which focused on nefarious activity rather than the outages, said hackers and cybercriminals are “moving beyond basic prompting” and using increasingly sophisticated AI techniques to devise and carry out their attacks. Techniques described in the report include using large language models and “agentic frameworks” to reverse engineer security patches, uncover vulnerabilities and exploit them.

The clustering of failures has also reached people who rely on these systems. Social media platforms have been abuzz with a mixture of annoyance, confusion, conspiracy theories and jokes about the string of outages. Jake Smith, a content creator who regularly posts videos about politics on platforms like Instagram and YouTube, said “To see several disruptions this close together, I think to a lot of people it’s made them realize how many parts of their lives they’ve handed over to these systems that we still, at this point, don’t fully understand.” Smith also said “I think it gives people a sense of helplessness to see all of it happening concurrently in a way.”

Democratic Sen. Mark Warner, co-chair of the U.S. Senate’s Cybersecurity Caucus, said Tuesday that “These disruptions should serve as a wake-up call. It feels like every day brings another reminder that AI is moving faster than our ability to manage its consequences.” Warner argued that the Trump administration had not shown enough interest in confronting these challenges, and said “based on what we know today, there is good reason to believe these disruptions are the unintended consequences of efforts to make our systems safer.”

Across the week, the disruptions reached email, workplace tools, flights, banking, consumer apps and government communications, and the organizations involved gave different accounts of what went wrong in each case. Together, the incidents illustrate the cascading effects that can erupt when a central player in technology, banking, travel or government encounters system errors that upend services. The reported causes differ, and no evidence presented links the string of failures to a cyberattack. Alongside the outages, AI is being used both to find flaws and, according to Google, to exploit them.