A review of the use of generative artificial intelligence (GenAI) tools by Department of Veterans Affairs (VA) staff has identified potential patient safety risks from a lack of safeguards and oversight. The review was conducted by the VA Office of Inspector General (OIG) between October 2025 and January 2026 and found that more than 15,000 VA staff members were using general-purpose GenAI chat tools authorized for use by the Veterans Health Administration (VHA) – VA GPT and Microsoft 365 Copilot Chat.

The reviewers identified broad staff engagement with the AI chat tools. An analysis of an internal prompt‑sharing application identified 135 prompts for the GenAI chat tools, 79 of which were clinical. The drafting of clinical notes and summarization of patient care were among the most common uses of the tools. The VA OIG notes that the tools were not specifically developed for clinical use, and while the VA provides clinical users with general training and resources, the VA does not centrally curate or evaluate prompts or the generative output, which may be applied to clinical decision making. The VA OIG notes that studies of genAI usage in medical settings found that prompt techniques can play a critical role in output errors that could impact diagnoses and care management if not corrected.

The Office of Management and Budget’s 2025 memorandum (Accelerating Federal Use of AI through Innovation, Governance, and Public Trust) requires all agencies to identify high-impact AI use and implement safeguards to manage risk. The VA did not identify the use of VA GPT and Copilot Chat as high-impact, and therefore, the required risk management actions did not apply.

The VHA had determined that Ambient AI Scribe was high-impact, which triggered safety requirements such as pre-deployment testing of the AI tool and providing human oversight before use. Ambient AI Scribe is a targeted clinical documentation tool that listens to clinical visits and drafts medical record notes. The VA-OIG said the tool had functionality similar to the clinical documentation prompts VA staff were using with VA GPT and Copilot Chat, which were not considered high-impact.

The VA OIG made three recommendations to the VHA regarding the use and assessment of GenAI chat tools: Evaluating these tools as high-impact, implementing the required safeguards, and integrating monitoring of AI-related risks into existing patient safety programs. The VHA concurred in principle with the recommendation to evaluate the tools as high -impact and concurred with the other two recommendations. The VHA has provided the VA OIG with an action plan, will develop guidance on the use of the GenAI chat tools, and is working on addressing the recommendations by April 2027.

As the use of GenAI tools in healthcare accelerates, concern is growing that sensitive patient data may be shared with publicly accessible chatbots, and that AI tools could generate output that puts patients at risk of harm or even death. Earlier this year, Health-ISAC and the Health Sector Coordinating Council Cybersecurity Working Group issued guidance on developing effective AI governance frameworks – Health-ISAC’s White Paper: Policies and Safeguards for a Safe Use of AI and the HSCC Health Industry AI Cyber Governance Framework Implementation Guide to help healthcare organizations create an effective AI governance and safeguards framework and responsibly use GenAI and LLMs while minimizing risk.

The post VA OIG Identifies Lack of Oversight of VA GenAI Chat Tools appeared first on The HIPAA Journal.

The Cybersecurity and Infrastructure Security Agency (CISA) has issued a Binding Operational Directive (BOD  26-04) establishing new deadlines for vulnerability remediation for federal civilian agencies. Defenders have long been struggling to keep on top of patching due to the frequency with which new vulnerabilities are identified, the pace of which has increased dramatically due to artificial intelligence.

According to the Verizon 2025 Data Breach Investigations Report, organizations were only able to fully remediate around 38% of vulnerabilities in CISA’s Known Exploited Vulnerability (KEV) Catalog in 2024. The 2026 DBIR report shows that the percentage of fully remediated vulnerabilities in 2025 fell to 26%, with a median resolution time of 43 days. Artificial intelligence has massively increased the pace of vulnerability discovery, defenders are becoming overwhelmed, and critical vulnerabilities are remaining unpatched for longer periods, increasing the window of opportunity for exploitation. CISA’s solution is to patch smarter, not harder.

CISA has released a new risk-based vulnerability remediation framework to help vendors assess vulnerabilities and prioritize patching effectively, concentrating their efforts on mitigating vulnerabilities in the most at-risk assets and addressing vulnerabilities that carry the greatest risk of exploitation.

CISA has determined that the greatest risk is associated with vulnerabilities with four characteristics:

1. Public exposure via the internet
2. The ability to fully automate exploitation
3. If the vulnerability gives an attacker full control of a system, and
4. Evidence of real-world exploitation (KEV inclusion)

Based on this framework, any vulnerability that meets all four criteria must be mitigated in the shortest possible timeframe – no more than 3 days. If the vulnerability is publicly exposed, is in the KEV, is automatable, and gives an attacker partial control of a system, the vulnerability must be remediated within 3 days. If the vulnerability gives an attacker full control of a system, following remediation within 3 days, a forensic triage is required to determine if the vulnerability has already been exploited.

New timelines have been provided for mitigating lower risk vulnerabilities of two weeks or two months, with the lowest severity vulnerabilities not requiring remediation until the next system upgrade. An analysis at one large civilian agency found that only 1% of vulnerabilities fell into the 3-day category, while 60% of vulnerabilities could be deferred unitl the next system upgrade. By following the new framework, organizations will be able to ensure that the most critical vulnerabilities are addressed first.

The new framework prioritizes mitigation of vulnerabilities at the network edge. While vulnerabilities in the network core may be high risk and under active exploitation, CISA generally does not observe threat actors compromising core networks through product vulnerabilities; they use living off the land (LOTL) techniques, which CISA says are best addressed through other means, such as system hardening, network segmentation, and implementing phishing-resistant multi-factor authentication.

The post CISA Instructs Federal Agencies to Adopt Risk-Based Approach for Vulnerability Remediation appeared first on The HIPAA Journal.

Patches have been issued to fix a critical vulnerability affecting Check Point Mobile Access, SSL VPN, Remote Access VPN, and Spark Firewalls, and a high-severity vulnerability in Google Chrome, both of which are being actively exploited in the wild.

Check Point Remote Access VPN Vulnerability

On June 8, 2026, the cybersecurity firm Check Point issued a security advisory about a critical authentication bypass vulnerability tracked as CVE-2026-50751 (CVSS 9.3), which has been actively exploited in zero-day attacks since May 7, 2026. Exploitation of the vulnerability accelerated over the weekend, with a few dozen organizations falling victim to attacks. In one attack, Check Point associated the post-exploit activity with a Qilin ransomware affiliate that has previously targeted vulnerabilities in other VPNs.

The vulnerability affects Check Point Mobile Access, SSL VPN, Remote Access VPN, and Spark Firewalls; however, only if deployments are configured to use the deprecated IKEv1 key exchange protocol. In vulnerable deployments, unauthenticated remote attackers can exploit a logic flaw in certificate validation, which allows them to establish a VPN connection without a valid password, bypassing authentication requirements.

Check Point also identified a second vulnerability while investigating the actively exploited zero day. The vulnerability is also associated with the deprecated IKEv1 key exchange, which can allow a man-in-the-middle attack on VPN site-to-site connections. The vulnerability is tracked as CVE-2026-50752, has a CVSS score of 7.4, and affects Security Gateways and Spark Firewalls. At the time of issuing the patch, there had been no known exploitation of the flaw.

Customers using the IKEv1 key exchange protocol have been advised to apply the security updates as soon as possible. If the hotfixes cannot be immediately applied, users should follow Check Point’s mitigation guidance detailed in the security alert. The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has added the vulnerability to its Known Exploited Vulnerability (KEV) Catalog and ordered all government agencies to secure their deployments by applying the security updates or mitigations within 3 days. or to discontinue use of the product.

Google Chrome Zero-day

Google has released an emergency patch to fix an actively exploited high-severity zero-day vulnerability in Google Chrome. The vulnerability, tracked as CVE-2026-11645, is due to an out-of-bounds read and write flaw in the Chrome V8 JavaScript engine. The vulnerability can be exploited by a remote attacker via specially crafted HTML pages. Successful exploitation allows the attacker to execute arbitrary code inside the web browser sandbox, exposing sensitive information or crashing Chrome.

Google is aware of an exploit for the vulnerability in the wild, and has rolled out updates for users in the Stable Desktop channel for Windows, Mac, and Linux Systems. Further information about the bug is being withheld until the majority of users have updated Chrome.

The post Check Point VPN and Google Chrome Vulnerabilities Under Active Exploitation appeared first on The HIPAA Journal.

The Health Sector Coordinating Council (HSCC) AI cybersecurity governance task force has published new guidance for healthcare CISOs and other leaders to help them establish cybersecurity governance frameworks for secure AI implementation.

Adoption of AI-based technologies in healthcare is progressing at a pace, with AI tools increasingly embedded into critical healthcare functions; however, these tools introduce new and often poorly understood cyber risks into already complex ecosystems. AI-specific cyber risks, such as data poisoning, model drift, and bias, can threaten successful implementation and HIPAA compliance, and the tools can create vulnerabilities that can be exploited by threat actors in attacks that impact patient privacy, safety, and care.

Healthcare organizations should implement a strong governance structure that integrates cybersecurity principles into the full AI product lifecycle, from assessment, design, development, deployment, and decommissioning of AI systems. The guidance can be used to implement a cybersecurity governance framework for identifying and mitigating AI-specific cyber risks associated with all AI technologies, from traditional machine learning systems to generative AI and agentic AI systems capable of autonomous action.

The AI Cyber Governance Framework Implementation Guide guidance establishes core AI cybersecurity governance objectives for enterprises, ecosystems, and third-party adoption scenarios, and includes AI cyber-specific industry best practices and protocols for secure data handling, model protection, continuous monitoring, and threat detection, including model evasion, model inversion, data leakage, and data poisoning. The guidance provides practical tools for organizing roles and responsibilities, inventory management, contractual language for vendor relationships, and includes a five-level AI autonomy framework and an AI-specific incident response playbook.

The 87-page guidance document is focused on establishing a governance framework for addressing AI-specific cybersecurity risks, and while the guidance covers clinical safety, ethics, and patient engagement when they intersect with cybersecurity risk, a broader AI governance program should be maintained for addressing the full spectrum of AI-related risks beyond cybersecurity, and should therefore be used in combination with existing organizational governance activities.

The playbook is part of a series of AI-specific documents for the healthcare industry, with previous publications including a guide for addressing supply chain risk. Further publications are expected in the coming months to address other healthcare-specific AI considerations.

The post HSCC Issues Guidance on Cyber Governance Frameworks for Secure AI implementation appeared first on The HIPAA Journal.

Healthcare organizations have embraced AI and are using AI agents to perform a range of functions, including handling IT support desk tickets, automating software workloads, authenticating data exchanges, and performing various security tasks. While there are clear benefits to be gained from using AI agents in healthcare, each new AI agent is a potential entry point for attackers, and a successful compromise could result in a devastating attack.

Each AI agent is given permissions to carry out its functions, and when AI agents are used to perform security functions, those permissions can be significant. Any attack that succeeds in compromising an AI agent will see the attacker gain those same permissions. For instance, an AI identity on a local machine may have access to the password manager, browser sessions, Secure Shell, and encryption keys. An AI agent could disclose admin credentials to an attacker, leading to a crippling attack with significant data theft.

To learn about AI deployments and integrations and how they are affecting identity security, the cybersecurity firm Semperis commissioned Censuswide to conduct a survey of 1,100 IT and IT security professionals across several industries, including healthcare. The survey confirmed that AI agents are being extensively deployed, which pose significant risks to identity infrastructure. Three-quarters of healthcare respondents believe that there will be AI-driven attacks on identity infrastructure, 69% believe that AI attackers will use identity systems to target their infrastructure, but only one-quarter of respondents think that they would be able to fully recover if an AI agent exposed administrative credentials.

On average, more than one-third of the healthcare workforce has at least one AI agent installed on a local machine that has permissions to access Secure Shell and encryption keys, and one in three healthcare respondents said they are using AI agents to handle security-related tasks, with 60% of respondents anticipating deploying AI agents for security tasks in the next 12 months.

According to Semperis, AI agents should be treated as non-human identities (NHIs) in the identity fabric; however, only 66% of respondents said AI identities were registered, authenticated, and authorized within the organization, and of those that do, almost half (48%) register, authenticate, and authorize them separately from human identities. While organizations may be applying security best practices such as the principle of least privilege for human identities, that is not always the case with AI identities, which are often overpermissioned.

“AI support agents are often overpermissioned in ways that may have unintended consequences — such as ‘helpfully’ reconfiguring security settings or granting access that can lock entire teams out of their identity systems or punch holes in corporate VPNs,” explained Semperis. As deployment of AI agents increases, so does the risk. Since AI agents often have the ability to do anything, it is vital to implement disciplined controls. While sufficient controls may not yet have been implemented, 90% of respondents said AI identity governance is a top security priority for the organization.

Semperis stresses that security controls need to be implemented to reduce risk, such as applying the principle of least privilege to AI identities, designating identity infrastructure, implementing backup and recovery controls, and segregating agent and human trust boundaries where appropriate. Organizations need to work on the assumption that AI identities will eventually be compromised, so they must therefore need to plan for that eventuality and ensure that they have the policies and procedures in place to allow them to rapidly respond and make a quick and full recovery.

“What’s striking isn’t just how quickly AI is being integrated into identity systems but how unprepared many organizations are to recover when things go wrong,” explained Grace Cassy, Partner, Ten Eleven Ventures. “Introducing AI at the identity layer offers operational advantages, but it must be accompanied by guardrails, observability, and recovery readiness. It’s a new dimension of an old question, really: Are you resilient enough to respond in the event of critical disruption?”

The Semperis State of Identity Security in the AI Era Report can be downloaded here.

The post Healthcare Orgs Lack Confidence in Ability to Defend Against an AI-incited Identity Breach appeared first on The HIPAA Journal.

Silent Ransom Group, a data theft and extortion group that targets law firms, healthcare organizations, and insurance and finance companies, is conducting a social engineering campaign posing as IT support workers. Silent Ransom Group (aka Luna Moth, Chatty Spider, UNC3753) is a financially motivated threat group that, as the name suggests, quietly infiltrates networks, exfiltrates sensitive data, and demands payment to prevent the stolen data from being publicly leaked or sold. The group does not use ransomware to encrypt files.

Silent Ransom Group has demonstrated a penchant for attacking U.S. law firms, although it has conducted attacks on other sectors such as insurance, finance, and healthcare, where the leaking of sensitive data can cause significant reputational harm and regulatory scrutiny. Silent Ransom Group has conducted phishing campaigns in the past, using social engineering techniques to trick employees into installing remote access software.

One such campaign involved phishing emails notifying the recipient about a subscription for a service that was about to incur a charge. The recipient was told that in order to prevent that charge, they must call the telephone number provided in the email. The call would be answered, and the user would be tricked into downloading remote access software, which was used to gain persistent access to the user’s systems. Data would be identified and exfiltrated, and a ransom demand would then be issued.

The latest campaign has been running since at least Spring 2026, according to a recent Federal Bureau of Investigation (FBI) Cyber Alert. A Silent Ransom Group actor poses as an employee of the victim’s IT department, contacting the victim over the telephone. In some cases, email will be used, requesting the victim contact the threat actor by phone.

Over the telephone, the user will be directed to grant access to a remote desktop session under the guise of fixing an IT issue. Should that attempt fail, the threat actor will arrange to visit the victim’s location in person to fix the issue. On an in-person visit, the threat actor will insert a storage device into the victim’s computer. The victim is told that they need to image the device or create a backup file to address potential impacts from the phishing email.

Once access is gained to a device, either physically or via a remote session, privileges are escalated minimally, and data is quickly exfiltrated, either to internal file-sharing platforms such as Google Drive or Microsoft OneDrive, or using WinSCP or Rclone. For the in-person visits, data is copied onto an external hard drive or USB drive.

In addition to raising awareness of the scam with employees, it is important to verify the identity of any individual attempting to gain physical access to company spaces. The FBI  has made several recommendations for improving defenses against Silent Ransom Group attacks in the alert, including strengthening authentication controls, informing employees about the scam, and strengthening physical security controls, including conducting checks of identification documentation before granting access to the facility.

The post Extortion Group Conducts Social Engineering Campaign Impersonating Victim’s IT Department appeared first on The HIPAA Journal.

Verizon has published its 2026 Data Breach Investigations Report, which shows that the healthcare sector continues to be targeted by cybercriminal groups. The sector is having to contend with sustained multi-vector attacks, including ransomware, unpatched vulnerabilities, and human error. Regardless of the cause, the attacks are putting patient privacy, safety, and care at risk.

Verizon tracked 1,492 healthcare incidents for its 2026 report, including 1,438 confirmed data disclosures, a majority of which were due to ransomware-driven system intrusions achieved through multiple attack vectors, including the exploitation of vulnerabilities (20%), phishing attacks (14%), stolen credentials (11%), and employee errors (11%). Threat actors are being given far too big a window of opportunity to exploit known vulnerabilities. Verizon found that in 2025, only 26% of critical vulnerabilities were fully remediated, with a median time for resolution stretching to 43 days. In healthcare, where complex legacy systems are the norm, the window of opportunity is greater, giving threat actors a wide attack window.

While external actors accounted for the majority of incidents, insider breaches remain common in healthcare. Internal actors were behind 19% of breaches. As Verizon notes, human error continues to be a chronic source of breaches. The human element was involved in 54% of incidents, including misconfigurations, misdirected communications, the loss/theft of unencrypted devices, and poor cyber hygiene.

The most common human-related cause of healthcare data incidents was misdelivery, which accounted for around 40% of incidents, followed by loss incidents at around 25%, and misconfigurations at around 20%. While greater investment in cybersecurity will help to address the 81% of breaches due to external actors, security awareness training plays an important part in preventing data breaches. Employees need to be made aware of security fundamentals and be taught the importance of practicing good cyber hygiene. Social engineering was the third main cause of healthcare breaches in 2025, the majority of which were due to phishing, followed by pretexting – these attack techniques need to be covered in depth in training courses.

Around 32% of healthcare data breaches involved third parties, so applying the security fundamentals internally is only part of the solution. Healthcare organizations must also ensure that they bake security into their contracts with business associates and suppliers. The proposed update to the HIPAA Security Rule, a final rule for which is expected at some point this year, will help to reduce the number of third-party breaches through more prescriptive security requirements for business associates and requiring greater vendor oversight by covered entities to ensure that security measures are implemented.

Each year, the number of real-world security incidents analyzed by Verizon continues to increase, and this year was no exception. The report covers more than 31,000 incidents, including 22,000 confirmed data breaches. GenAI tools are increasingly used by threat actors to accelerate and increase the volume of attacks. GenAI is being used at various stages of the process, including choosing targets, researching vulnerabilities, developing malware, gaining a foothold in networks, and making their campaigns more efficient and effective.

Overall, across all sectors, system intrusions continue to be the top breach pattern, with ransomware the primary driver. Last year, stolen credentials were the top entry point, but this year, this long-standing common attack vector has been overtaken by vulnerability exploitation. This is the first time in Verizon’s 19 years of producing its DBIR reports that vulnerability exploitation has topped the list. Verizon attributes this change to the use of AI by attackers, which has helped them accelerate the time to exploit known vulnerabilities. Defenders now have far less time to remediate vulnerabilities. While the time from disclosure to exploitation used to be measured in months, vulnerabilities are now being exploited in hours.

Ransomware continues to be a key driver of intrusions. Ransomware-related intrusions grew in volume again and now account for 48% of all breaches, up from 44% last year, although the percentage of victims paying a ransom is decreasing, as is the median ransom payment. In the past year, 69% of victims chose not to pay the ransom, and the median ransom payment fell from $150,000 to $139,875.

Awareness about email phishing has grown, making this attack technique less successful. Threat actors have responded by pivoting to mobile-centric social engineering techniques such as text messages (smishing) and voice phishing (vishing), where the success rate is 40% higher than traditional email phishing. Verizon warns that the easy availability of GenAI tools is creating a significant data security risk. Employees are increasingly using genAI tools without the knowledge or approval of the IT department. The massive increase in shadow GenAI use creates a significant risk of data exfiltration through unapproved platforms. This is particularly concerning for regulated sectors such as healthcare.

“While the velocity of cyber threats—driven by AI and faster vulnerability exploitation—is increasing, the foundational principles of security and strong risk management remain the most effective defense,” said Daniel Lawson, SVP Global Solutions, Verizon Business. “The DBIR reinforces that these fundamentals still hold as organizations strive for resilience.”

The post Verizon: Healthcare Sector Facing Sustained, Multi-vector Attacks appeared first on The HIPAA Journal.