Germany’s Federal Office for Information Security (BSI) has raised its assessment of the country’s IT security situation from “tense” to “tense to critical,” warning that artificial intelligence is helping attackers find weaknesses and develop exploits more quickly. The agency announced the change in its 2026 annual report, published on November 19 and covering July 2025 through June 2026.
The report introduces the BSI Cyber Index, which averaged 170 points during the period, 50% above the previous reporting period. Reuters reported that the index reached 238 points in the second quarter of 2026, compared with 124 at the same point in 2025. The BSI also recorded a 117% increase in its attack-surface score and a 48% rise in exploitation attacks.
The gap between disclosure and exploitation narrowed. The BSI measured an average of 25 days between a vulnerability becoming known and its first known exploitation, down from 53 days in the previous period. In the second quarter of 2026, that average dropped to eight days. The agency said around 160 new vulnerabilities were disclosed each day, 35% more than in the previous period.
The report warns that advanced AI models can search for weaknesses, generate exploit code and chain flaws into attack paths. It also describes incidents in summer 2026 in which AI agents escaped test environments and carried out exploitation attacks against unrelated third parties. The BSI’s warning echoes Google’s report on AI agents automating real-world exploits.
The BSI said many organizations and consumers still neglect basic protections. Fewer than half of consumers use measures such as strong passwords. In the agency’s CyberRiskCheck, participating organizations met an average of only 55% of minimum cybersecurity requirements, even though 90% of surveyed companies rated their own security as good or very good.



