AI cyberattacks make software vulnerabilities the top breach trigger, Verizon Data Breach Report warns

Software Vulnerabilities Become the Leading Entry Point

The exploitation of software vulnerabilities has emerged as the dominant threat vector in corporate cybersecurity, accounting for 31% of all confirmed data breaches according to Verizon's 2026 Data Breach Investigations Report1

1

. This marks the first time in nearly two decades that vulnerability exploitation has overtaken stolen credentials, which have dropped to just 13% of reported incidents1

1

. The Verizon Data Breach Report analyzed over 31,000 security incidents across 145 countries, revealing a fundamental shift in how threat actors gain initial access to corporate systems2

2

.

AI Cyberattacks Accelerate Exploitation Speed

AI-powered hackers are leveraging artificial intelligence to accelerate the discovery and weaponization of known software flaws, dramatically shrinking the window available for defenders to patch their systems from months to mere hours1

1

. Verizon warned that AI is being used by threat actors across different stages of attacks, including targeting, initial access, vulnerability research, malware deployment and other tooling2

2

. In the median case, threat actors researched or used AI assistance in 15 documented techniques, while some used it across 40 or 50 techniques2

2

. Despite this growing risk, only 26% of critical vulnerabilities were fully remediated throughout 2025, with the median time organizations took to apply patches jumping to 43 days1

1

.

Mobile Phishing Attacks Outperform Email Scams

Mobile devices have become a more dangerous attack vector than email, with phishing simulations showing that text messages and voice calls achieve 40% higher click rates than traditional email phishing1

1

. Phone-centric tactics, including voice and text-message attacks, are becoming more effective as attackers increasingly target mobile-centric communication channels where users are less suspicious2

2

. The human element was still involved in 62% of all breaches, while AI-fueled social engineering represented 16% of all breaches2

2

.

Ransomware and Third-Party Compromises Surge

Ransomware was present in nearly half of all breaches at 48%, up from 44% the previous year, though 69% of victims refused to pay1

1

. Supply chain attacks have also grown substantially, with third-party involvement in breaches increasing by 60% year-over-year, reaching 48% of total breaches1

1

2

2

.

Unauthorized AI Tools Create Data Leakage Risks

Nearly half of all employees, or 45%, now use AI tools at work, representing a significant increase from just 15% the previous year1

1

. However, 67% of these workers access generative AI platforms through unauthorized personal accounts rather than approved corporate channels, creating what experts call "Shadow AI"1

1

. This has become the third most common cause of non-malicious data leakage, putting company secrets at significant risk of unintended exposure.

Defensive Cybersecurity Strategies Must Evolve

Verizon's report argues that AI is, for now, mostly making familiar attacks faster, cheaper and more scalable, with less than 2.5% involving less common techniques2

2

. "AI's primary impact is currently operational: automating and scaling techniques defenders already know how to detect, not yet unlocking these novel or rare attack surfaces," the report states2

2

. Verizon chief information security officer Nasrin Rezai emphasized the need to fight AI with AI, incorporating these tools into software development life cycles, testing processes and cyber defense processes at unprecedented scale4

4

. The report does not cover data from Mythos, a new AI model announced on April 7 as part of Anthropic's "Project Glasswing," which has raised widespread cybersecurity concerns due to its high-level coding skills and ability to identify vulnerabilities4

4

.