started · updated
AI accelerates cyberattack speed and creates new security vulnerabilities
The rapid advancement of artificial intelligence is fundamentally altering the cybersecurity landscape, creating both new defensive capabilities and unprecedented risks. The Bank for International Settlements (BIS) has issued a warning through its Financial Stability Institute, noting that the window between the discovery of software vulnerabilities and their exploitation has shrunk from weeks to mere minutes due to AI-driven automation.
Financial institutions are particularly vulnerable, as traditional patching schedules may no longer be sufficient to counter autonomous exploitation. Regulators in the United Kingdom, Germany, and Hong Kong are urging faster response times and enhanced operational resilience. This shift is echoed by Visa, which released a whitepaper on frontier AI and cyber resilience to help organizations transition toward continuous, intelligence-driven security operations.
Beyond automated attacks, researchers have identified new categories of risk. Noma Labs highlighted “workflow identity hijacking,” where AI agents can be manipulated into performing privileged actions via unauthenticated inputs. Additionally, the integration of AI agents into business operations introduces “non-human insider risks,” characterized by machine-speed actions, excessive privileges, and potential influence from untrusted data inputs. Google Threat Intelligence Group further reported that both state-linked and criminal actors are increasingly incorporating agentic AI into their attack frameworks to scale and accelerate their operations.
Entities
Anthropic · Bank for International Settlements · Claroty · Financial Stability Institute · Google Threat Intelligence Group · Noma Labs · Tata Elxsi · Visa
Claims
What the coverage asserts, and how many sources carry each claim.
- [● 3 SOURCES] AI has reduced the time between software vulnerability discovery and exploitation from weeks to minutes. bitnewsbot.com · www.ad-hoc-news.de
- [○ 1 SOURCE] State-linked and criminal hackers are using agentic AI and automation to scale attacks and bypass defenses. www.cybersecuritydive.com
- [○ 1 SOURCE] AI agents create non-human insider risks through persistent access, excessive privilege, and untrusted inputs. cybernoz.com
- [● 3 SOURCES] Frontier AI enables autonomous discovery and exploitation of software vulnerabilities. bitnewsbot.com · www.ad-hoc-news.de
- [○ 1 SOURCE] Visa released a whitepaper addressing cyber resilience in the era of frontier AI. businessnewsthisweek.com
- [○ 1 SOURCE] AI workflows can suffer from workflow identity hijacking, where unauthenticated users trigger privileged actions. www.csoonline.com