The average cost of downtime has reached $600 billion across the Global 2000, a 50% increase in just two years, according to a newly published report from Splunk.
Produced in partnership with Oxford Economics, Splunk’s “The Hidden Costs of Downtime” report highlights the rising financial impact of cyber incidents, outages, and breaches, including a growing set of “hidden costs” that organizations often overlook.
Downtime costs rise across the Global 2000
According to the study, downtime has become increasingly unavoidable due to internal vulnerabilities, escalating cyber threats, growing AI complexity, and reliance on third-party providers.
Together, these issues have contributed to an average of 60 downtime incidents per year.
“The aggregate cost of downtime for the Global 2000 is now $600 billion — up 50% in just two years. That works out to an average of $300 million per organization each year. The expenses add up quickly: Every minute of downtime costs $15,000, or over $900,000 an hour,” the report said.
Hidden costs intensify after breaches and outages
Beyond the direct financial impact, Splunk found that the “hidden costs” associated with downtime have become significantly more severe.
Key findings include:
- Publicly disclosing a data breach was identified as the top hidden cost, with 71% of respondents rating it as “very” or “prohibitively disruptive,” up from 23% in 2024.
- The number of respondents who said deploying large numbers of personnel to recover from an outage was “very” or “prohibitively disruptive” rose to 68% in 2026, up from 24% in 2024.
“Public disclosure of a data breach is more than three times as disruptive, and losing customers is more than twice as disruptive as in 2024,” the report emphasized.
“What’s changed? For starters, things got personal for executives. New SEC rules make leaders like CISOs and CFOs responsible for reporting security breaches. Failure to do so correctly could cost them financially and legally. Recent high-profile cyberattacks also demonstrated the serious consequences of a breach.”
AI tools reduce risk but introduce new failure points
Splunk also noted that lower switching barriers for customers have amplified the business impact of downtime.
“Users now face lower technical hurdles to change services, so patience for poor performance has worn thin. In competitive B2C markets, downtime and service degradation
send customers fleeing to competitors. In B2B and regulated industries, they threaten future contracts.”
Additional findings from the report include:
- Stock prices decline by an average of 3.4% following a downtime event.
- 49% of technology executives said shareholders have raised concerns about downtime within the past 12 months.
- Companies spend a median of $24.5M annually on AI tools that prevent and respond to downtime.
- Among organizations using AI tools, 56% of users said AI has reduced the overall risk of downtime.
Shadow AI creates new security and operational concerns
Despite the generally positive perception of AI’s role in reducing risk and preventing downtime, Splunk’s report also pinpointed a growing concern: the rise of shadow AI.
According to the study, 66% of respondents said employees are using unapproved AI tools, commonly referred to as “shadow AI,” to assist with their work, introducing additional security and operational risks into enterprise environments.
“In my view, these figures reflect where we are in the AI adoption curve,” said Hanlin Fang, vice president of product management at Splunk.
Human oversight remains critical to AI resilience
In the report, Fang emphasized that organizations that successfully minimize AI-related downtime are not necessarily those with the most sophisticated technology.
Rather, they are the organizations that have “humans in control,” along with continuous monitoring and rapid intervention when outcomes begin to drift.
The report also found that many technology executives have already experienced downtime caused by AI-related issues.
Percentage of technology executives who experienced downtime caused by the following AI-related issues:
- Incorrect AI-driven automation – 50%
- Model drift – 50%
- Integration bugs from embedding AI into production systems – 29%
- Adversarial attacks – 26%
Earlier this year, Channel Insider covered RSAC 2026, where AI dominated discussions as vendors unveiled new security tools and capabilities tied to the emerging technology. Learn more about how cybersecurity providers are approaching AI and how the technology is reshaping enterprise security as we know it.