{"id":182385,"date":"2026-01-28T11:52:00","date_gmt":"2026-01-28T16:52:00","guid":{"rendered":"https:\/\/testing.news-you-need.com\/index.php\/2026\/01\/28\/ai-tools-break-quickly-underscoring-need-for-governance\/"},"modified":"2026-01-28T12:45:09","modified_gmt":"2026-01-28T17:45:09","slug":"ai-tools-break-quickly-underscoring-need-for-governance","status":"publish","type":"post","link":"https:\/\/testing.news-you-need.com\/index.php\/2026\/01\/28\/ai-tools-break-quickly-underscoring-need-for-governance\/","title":{"rendered":"AI tools break quickly, underscoring need for governance"},"content":{"rendered":"

AI tools break quickly, underscoring need for governance<\/a><\/p>\n

https:\/\/www.cybersecuritydive.com\/news\/ai-vulnerabilities-governance-zscaler\/810718\/<\/a><\/p>\n

Publish Date: 2026-01-28 11:52:00<\/a><\/p>\n

Source Domain: www.cybersecuritydive.com<\/a><\/p>\n

Author: <\/a><\/p>\n

Using an unordered list, summarize the following article with between 4 and 8 key points. <\/p>\n

Listen to the article
\n 3 min<\/p>\n

This audio is auto-generated. Please let us know if you have feedback.<\/p>\n

Dive Brief:<\/p>\n

Companies\u2019 AI tools remain highly vulnerable to cyberattacks, even as enterprises race to use them in more ways, the security firm Zscaler said in a threat report published on Tuesday.
\nEnterprises are also feeding AI tools vastly more data, the report found, \u201cwhich paints an expanding target on AI platforms for cybercriminals across the globe.\u201d
\nZscaler recommended organizations focus on visibility, real-time defense and consistent governance controls.<\/p>\n

Dive Insight:
\nOne of the most striking findings in Zscaler\u2019s report concerns how brittle many AI systems are. \u201cThey break almost immediately,\u201d researchers wrote. \u201cWhen full adversarial scans are run, critical vulnerabilities surface within minutes \u2014 and sometimes faster.\u201d During Zscaler\u2019s red-teaming exercises in 25 corporate environments, it took a median of 16 minutes for an AI system to experience its first major failure, and by 90 minutes, 90% of systems had failed. In one case, it took only a single second for a system to fail.
\nResearchers observed failures in categories including biased and off-topic responses, failed URL verifications and privacy violations. \u201cModels can still be coerced into exposing sensitive data or participating in harmful workflows,\u201d the report warned.
\nIn 72% of corporate environments, Zscaler\u2019s first test of an AI system uncovered a critical vulnerability.
\nThe lesson for CISOs, the report said, is that \u201ccritical risk is present from day one, even in mature environments,\u201d and security personnel need to constantly test their systems and apply strict governance protocols.
\nAt the same time, Zscaler\u2019s analysis of nearly one trillion AI data transactions in its cloud environment in 2025 revealed some promising signs. Companies\u2019 security policies blocked roughly 40% of all attempted AI transactions, a phenomenon that Zscaler said reflected \u201cgovernance in action \u2026 as leaders balance the tradeoff between innovation speed and risk tolerance.\u201d
\nThe 989.3 billion AI transactions that Zscaler observed in 2025 represented a 91% increase from 2024, with the company tracking activity from more than 3,400 different AI tools.
\nThe U.S. accounted for roughly 38% of the transactions, followed by India (14%) and Canada (5%). And the finance and manufacturing sectors led the way in using AI for the third year in a row, representing 23% and 20% of AI transactions, respectively, in 2025.<\/p>\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

AI tools break quickly, underscoring need for governance https:\/\/www.cybersecuritydive.com\/news\/ai-vulnerabilities-governance-zscaler\/810718\/ Publish Date: 2026-01-28 11:52:00 Source Domain:…<\/p>\n","protected":false},"author":1,"featured_media":182386,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"fifu_image_url":"https:\/\/imgproxy.divecdn.com\/JkbV_h7Jd9EsgKEfO74cnxYS7DvgOoTZzBgh5gLfim4\/g:ce\/rs:fit:770:435\/Z3M6Ly9kaXZlc2l0ZS1zdG9yYWdlL2RpdmVpbWFnZS9HZXR0eUltYWdlcy0xNTQ2MjIzNDYyLmpwZw==.webp","fifu_image_alt":"","footnotes":""},"categories":[15],"tags":[26,27],"class_list":["post-182385","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-cybersecurity","tag-ai","tag-vulnerability"],"_links":{"self":[{"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/posts\/182385"}],"collection":[{"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/comments?post=182385"}],"version-history":[{"count":1,"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/posts\/182385\/revisions"}],"predecessor-version":[{"id":182387,"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/posts\/182385\/revisions\/182387"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/media\/182386"}],"wp:attachment":[{"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/media?parent=182385"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/categories?post=182385"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/testing.news-you-need.com\/index.php\/wp-json\/wp\/v2\/tags?post=182385"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}