Rushed AI Security Patches Trigger Global Tech Outages
US officials and cybersecurity experts have expressed apprehensions that an increasing urgency to address software vulnerabilities identified by artificial intelligence may be linked to a spate of recent technological disruptions, as reported. In the preceding week, numerous prominent entities, such as Microsoft, JPMorgan, the Federal Aviation Administration, and the US National Security Agency, have disclosed interruptions to their computer systems and online services. Microsoft services, including Outlook, Teams, and Copilot, experienced a significant disruption last Friday. The outage impacted thousands of users in the US before extending into the Asia-Pacific business hours as the company faced challenges in addressing server-related issues. The disruption also impacted government agencies in Australia, Japan, and the US, as well as firms like OpenAI and Anthropic. It represented the second significant interruption to Microsoft’s 365 services in the current year.
The outage persisted for over 24 hours, rendering users incapable of sending emails, accessing calendars, or conducting online meetings via Outlook, Teams, Copilot, and Exchange Online. Businesses, government agencies, airlines, and financial institutions were among those impacted. Downdetector recorded over 7.4 million reports throughout the incident, with the volume of reports peaking at nearly 20,000 per minute. Amid speculation regarding potential cyberterrorism, Mark Warner, vice chair of the US Senate Select Committee on Intelligence, indicated that the disruptions were more plausibly associated with organisations hastily addressing software vulnerabilities. The report quoted him as saying that these incidents were not due to a cyber attack, foreign interference or some co-ordinated effort. “What appears to be happening instead is a consequence of how quickly artificial intelligence is changing the cybersecurity landscape,” he said.
AI tools have transformed the methodology employed by security experts in the identification of software vulnerabilities. They possess the capability to analyse vast quantities of computer code rapidly, identifying bugs and potential solutions at a pace that significantly outstrips human ability. This has initiated a competitive environment among organisations to detect and rectify vulnerabilities prior to their exploitation by hackers, according to the report. Warner indicated that the escalating capabilities of AI are exerting significant pressure on corporations and governmental entities to address security vulnerabilities without delay, often necessitating the implementation of multiple updates concurrently. The US National Security Agency also seems to have been impacted by the recent disruptions. The signals intelligence agency reportedly encountered issues with a classified internal communications network on Wednesday, as reported.
The report cited Anne Neuberger, a former deputy national security adviser and former NSA director, stating that even agencies tasked with national security were not immune to such issues. “No one is immune as organisations rush to secure systems one step ahead of attackers,” she said. “In some cases, applying multiple patches will lead to outages. I suspect that is what occurred here.” The NSA said it had experienced “temporary degradation during scheduled routine maintenance”. The recent incidents have impacted multiple sectors and highlighted the reliance of businesses and public institutions on digital systems. Last week, email services across businesses and institutions in over 70 countries faced disruptions due to a nearly 24-hour outage of Microsoft 365. On Friday, the FAA suspended flights nationwide due to what it characterised as instability in its “legacy equipment.”









