Verizon report reveals AI is accelerating social engineering attacks and data breaches

Verizon Report Warns of AI-Fueled Social Engineering Surge | PYMNTS.com

By completing this form, you agree to receive marketing communications from PYMNTS and to the sharing of your information with our sponsor, if applicable, in accordance with our Privacy Policy and Terms and Conditions. That is one of the central findings in Verizon's 2026 Data Breach Investigations Report, which analyzed more than 31,000 real-world security incidents, including more than 22,000 confirmed data breaches across 145 countries. The report found that threat actors are using generative AI across different stages of attacks, including targeting, initial access, vulnerability research, malware development and other tooling. The finding does not mean every scam is suddenly new. Verizon's report argues that AI is, for now, mostly making familiar attacks faster, cheaper and more scalable. In the median case, threat actors researched or used AI assistance in 15 documented techniques, while some used it across 40 or 50 techniques. Most AI-assisted malware and tooling was tied to already known attack methods, while less than 2.5% involved less common techniques. That makes the threat practical for banks, payments companies and digital commerce firms. AI can help a less sophisticated criminal write better phishing messages, automate parts of reconnaissance, test targets faster or impersonate trusted contacts more convincingly. Verizon also found that the human element was present in 62% of breaches, while social engineering represented 16% of all breaches. Phone-centric tactics, including voice and text-message attacks, are becoming more effective, with simulated mobile-centric attacks producing median click rates 40% higher than email. The broader fraud picture is not limited to AI. Exploitation of vulnerabilities became the most common initial access vector for breaches, rising to 31%, while credential abuse fell to 13%. Ransomware appeared in 48% of breaches, up from 44% the prior year, although 69% of ransomware victims did not pay. Third-party involvement also rose sharply, reaching 48% of total breaches. The implication is that companies do not need to throw out their security playbooks, but they do need to speed them up. Verizon put it this way: "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." For financial services firms, that means the AI fraud challenge is not only about exotic deepfakes or autonomous hackers. It is about more-convincing scams, more pressure on help desks, more attacks against mobile channels and more criminals using automation to find weak spots before defenders close them.

AI-related data breaches surpass stolen credentials in cyber incidents, Verizon report says

WASHINGTON, May 19 (Reuters) - AI-detected vulnerabilities surpassed incidents of stolen credentials in data breaches last year, according to an annual report from Verizon relating to industry security incidents. Verizon said in a review of more than 31,000 incidents that 31% of all breaches started with vulnerability exploitation in an AI world. It warned that AI was being used by threat actors "to accelerate the time to exploit known vulnerabilities, shrinking the window for defense from months to mere hours." The annual report that reviews a wide range of industry data shows hackers are using generative AI to help at all stages of attacks "including targeting, initial access, and development of malware and other tools." The report said 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." But it added that assessment might be obsolete as AI continues to advance rapidly. The report said threat actors typically researched or used AI assistance in 15 different techniques, with some using as many as 50. The report does not cover data from Mythos, a new AI model that has raised widespread cybersecurity concerns. Mythos, ?announced on April 7, is being deployed as part of Anthropic's "Project Glasswing," a controlled initiative under which select organizations are permitted to use the unreleased Claude Mythos Preview model for defensive cybersecurity purposes, including Verizon. Mythos' skill in coding at a high level has given it a potentially unprecedented ability to identify cybersecurity vulnerabilities and devise ways to exploit them, according to experts. Verizon chief information security officer Nasrin Rezai said it was critical to address the growing threats. "We need to fight AI with AI. We need to incorporate them into our practices," Rezai told Reuters. "We need to bring them into our software development life cycle, in our testing processes, in our cyber defense processes at a scale that we have never done before." (Reporting by David Shepardson; Editing by Bernadette Baum and Hugh Lawson)