< OLDERNEWER >
May 23, 2025
The 2025 Verizon Data Breach Investigations Report highlights how people, ransomware, vulnerable systems, third parties, stolen credentials, unmanaged devices, and emerging AI use continue to shape cybersecurity risk. In this Hotman Group security awareness training, Tanya Wade breaks down several of the report’s key findings and the practical security behaviors organizations should reinforce.
This training reviews selected findings from the 2025 Verizon Data Breach Investigations Report, including human involvement in breaches, ransomware, vulnerability exploitation, third-party risk, credential exposure, unmanaged devices, AI use, and practical steps organizations can take to reduce risk.
The 2025 Verizon DBIR reinforces a familiar cybersecurity lesson: security failures rarely come from one isolated problem.
People, credentials, vulnerabilities, third parties, devices, access, ransomware, and newer technologies such as AI all create potential paths for attackers.
Organizations can reduce those risks by strengthening security awareness, patch management, multifactor authentication, vendor oversight, device management, secrets management, least privilege, network segmentation, and governance around AI use.
The Verizon Data Breach Investigations Report, commonly called the DBIR, is an annual analysis of cybersecurity incidents and confirmed data breaches.
The 2025 report analyzed more than 22,000 security incidents and more than 12,000 confirmed breaches across 139 countries.
Rather than focusing on one type of attack, the report examines patterns across threat actors, attack methods, assets, industries, vulnerabilities, ransomware, third parties, credentials, and other factors that help organizations understand how breaches are actually occurring.
People continue to play a significant role in cybersecurity incidents.
In the 2025 DBIR, Tanya highlights that 60% of the breaches analyzed involved a human element.
That can include phishing, stolen credentials, mistakes, misdirected information, and other situations where human behavior creates or contributes to an attack path.
Security awareness is not about expecting employees to become cybersecurity experts. It is about helping people recognize risk and know when to stop, verify, and report.
Ransomware remained a major theme in the 2025 report.
Tanya highlights that ransomware appeared in a larger share of breaches than in the previous year's report, with smaller organizations especially affected.
The takeaway for employees and organizations is not that ransomware is inevitable. Strong credentials, awareness, patching, backups, access controls, endpoint security, monitoring, and incident-response readiness can all reduce ransomware risk and limit potential impact.
Attackers do not always need a sophisticated new technique. Known weaknesses can provide an effective path into an environment when they remain exposed.
The training highlights increased exploitation of vulnerabilities as an initial access method, including attacks affecting edge devices and VPN-related infrastructure.
This makes patch management more than an IT maintenance task.
Organizations need processes for identifying vulnerabilities, evaluating their severity and relevance, prioritizing remediation, assigning ownership, and tracking whether fixes are actually completed.
Organizations increasingly rely on outside providers to store data, process information, operate technology, deliver cloud services, and support critical business processes.
That means another organization's security can become part of your organization's risk.
The 2025 DBIR data discussed in the training showed third-party involvement increasing substantially compared with the prior year.
The practical response is ongoing vendor management: understand critical dependencies, evaluate security controls during onboarding, identify risk, establish expectations, and periodically reassess whether the vendor continues to meet the organization's needs.
Credentials can allow attackers to look like legitimate users.
Passwords may be stolen through phishing, malware, compromised personal devices, previous breaches, credential reuse, or markets where attackers buy and sell access.
Once attackers have valid credentials, a missing second layer of authentication can make access significantly easier.
That is one reason multifactor authentication remains an important control: a stolen password alone should not automatically provide access to sensitive systems or information.
Employees may access company email, files, applications, and other sensitive information from personal phones, tablets, or computers.
If the organization has no visibility into those devices, it may not know whether they are patched, protected, encrypted, appropriately configured, or already compromised.
Organizations that allow bring-your-own-device access should therefore establish appropriate BYOD policies, access requirements, device-management practices, and security controls based on the sensitivity of the information being accessed.
Credentials, API keys, tokens, and other secrets can create significant exposure when they are embedded directly in code or stored in public repositories.
Attackers can search exposed repositories for credentials that may provide access to systems, cloud environments, APIs, or data.
Organizations should minimize hard-coded credentials, maintain appropriate secrets-management processes, and respond quickly when exposed secrets are discovered.
AI tools can improve productivity, but they also create new opportunities for information to leave the organization's controlled environment.
Employees may enter company data, customer information, internal documents, code, or other sensitive material into AI platforms without fully understanding how that information will be stored, processed, protected, or used.
The training emphasizes the importance of clear AI policies, employee education, approved-use guidance, and appropriate limitations on what information can be entered into AI systems.
AI can be useful without becoming unmanaged risk. The difference is governance. Learn more about building AI governance into the broader risk program.
Multifactor authentication adds another layer of verification beyond a password.
That becomes particularly important when credentials are stolen, reused, exposed, or purchased by an attacker.
Organizations should consider MFA broadly across systems and accounts rather than leaving avoidable gaps where a stolen password can still provide direct access.
Least privilege means giving people only the access they need to perform their roles.
If an account is compromised, unnecessary permissions can give an attacker additional systems, data, and actions to exploit.
Regularly reviewing access and limiting unnecessary privileges can therefore help reduce the impact of stolen credentials and account compromise.
Network segmentation helps prevent an attacker who compromises one part of an environment from moving freely throughout the rest of it.
Segmentation does not prevent every incident, but it can help contain malicious activity and reduce the amount of damage one compromised account, system, or device can cause.
The Verizon Data Breach Investigations Report is an annual analysis of security incidents and confirmed data breaches that identifies common attack patterns, threat actors, vulnerabilities, affected assets, and other cybersecurity trends.
Themes discussed in this training included human involvement, ransomware, vulnerability exploitation, edge devices, third-party risk, stolen credentials, unmanaged personal devices, exposed secrets, AI use, and industry-specific threats.
Attackers frequently rely on phishing, stolen credentials, mistakes, social engineering, and other human interactions. Security awareness helps employees recognize suspicious situations and respond appropriately.
Organizations rely on vendors for data, technology, infrastructure, and critical services. A weakness at a third party can therefore create security, operational, privacy, or business risk for the organizations that depend on it.
Organizations can reduce credential-related risk through strong authentication, broad use of MFA, appropriate access controls, secrets management, device security, reducing credential reuse, and training employees to recognize credential-theft attempts.
Employees may unintentionally expose sensitive information when using AI platforms. Clear policies, approved tools, data-handling rules, access controls, and employee education help organizations use AI while reducing unmanaged risk.
This article is based on a Hotman Group security awareness training presented by Tanya Wade covering selected findings from the 2025 Verizon Data Breach Investigations Report.
The training focuses on practical awareness topics including human involvement in breaches, ransomware, vulnerability management, third-party risk, stolen credentials, unmanaged devices, secrets management, AI usage, multifactor authentication, least privilege, and network segmentation.
Reports such as the DBIR can help organizations understand how attacks are changing, but awareness only creates value when those lessons are translated into practical controls, processes, employee behaviors, and risk decisions.
Hotman Group is a cybersecurity and Cyber GRC professional-services firm that helps organizations diagnose, design, build, remediate, implement, operate and mature cybersecurity programs.
HG helps organizations connect cybersecurity risks to practical governance, controls, operating processes, and business strategy so security programs continue to mature as threats, technology, and business requirements change.