Issue 8
China-linked hackers use VMware ESXi zero-days to escape virtual machines
According to reports, China-linked hackers are breaking out of virtual machines using two ESXi zero-day flaws. One article says the flaws were likely exploited a year before public disclosure.
Who is affected and how
- security leader
- Given the long period of exploitation before disclosure, a retrospective investigation is also warranted.
- security operations
- ESXi hosts need to be patched and searched retroactively for signs of earlier compromise.
FBI warns North Korean group is using malicious QR codes in attacks
An FBI warning issued on 9 January 2026 says a North Korean cyber threat group is using malicious QR codes in its attacks. The warning was covered by 13 articles.
security leaderQR-code deception should be added to the threat model and to user awareness materials.citizenA QR code of unknown origin should not be scanned without verifying its source.vulnerability · United StatesCISA flags Microsoft Office and HPE OneView vulnerabilities as actively exploited
On Wednesday, 7 January 2026, CISA added two vulnerabilities to the KEV catalog: one in Microsoft Office and one in HPE OneView. According to Malwarebytes, the older PowerPoint flaw was previously…
security leaderBecause old flaws are being exploited in practice, an inventory of software past end of support is also important.security operationsAffected HPE OneView and Office instances must be checked and patched urgently.regulation · United KingdomUK launches government cyber action plan and £210m cyber unit
In early January 2026, the UK government unveiled the Government Cyber Action Plan, which includes a new government cyber unit backed by £210m in funding. According to reports, the plan aims to…
security leaderIn the public sector, replacing legacy systems and enforcing accountability mean stricter requirements.public-sector decision-makerThe action plan sets out a funding and accountability model for public-sector cyber resilience that other governments can measure themselves against.report · United StatesCISA's KEV catalog grew about 20% in 2025, passing 1,480 entries
The catalog of known exploited vulnerabilities (KEV) maintained by CISA grew by about 20% in 2025, according to an analysis by Cyble. The number of entries exceeded 1,480.
security leaderThe growing catalog points to a rising number of exploited flaws, which matters when planning patching capacity.security operationsThe KEV list is one of the most important external inputs for setting patching priorities.Hungary
More key stories of the week
Baltic undersea cable damage: Finland seizes ship, Latvia finds no link
An undersea cable connecting Finland and Estonia was reportedly damaged, and Finland seized a ship whose crew came from four countries, including Russia. According to another report, the Latvian…
security leaderThe exposure of undersea communications infrastructure belongs in the risk assessment for critical infrastructure.public-sector decision-makerPolice say some Baltic cable incidents are sabotage while others are accidents or subject to ongoing proceedings, so the cause must be established case by case.incident · globalLedger customers affected by data breach at third party Global-e
Ledger customers were affected by a data breach at an external provider, Global-e, which was reported on 5 January 2026. The 12 articles discuss the leak that occurred through the provider chain.
security leaderThe data handling of external commerce and payment providers is part of an organisation's own risk management.business leaderPartner contracts should set out accountability for incident notification and data protection.citizenLedger customers should expect phishing attempts that use their data.jelentés · ÁzsiaTaiwan's NSB reports rise in Chinese cyberattacks, citing fake news sites and botnets
According to a report by Taiwan's National Security Bureau (NSB), the number of cyberattacks attributed to China has risen. Another report says the NSB cites fake news sites and bots as tools of…
public-sector decision-makerThe official report illustrates handling cyberattack and disinformation threats together.journalist and researcherIndependent verification of the fake news sites and botnet activity named by the NSB may be a task for the profession.incident · United StatesCovenant Health ransomware attack affects more than 284,000 people in Maine
According to reports, the number of people affected by the ransomware attack on Covenant Health rose by 5,980% from the initial disclosure. Another report puts the number of affected people in Maine…
security leaderThe later multiplication of the affected count shows that early disclosures are often underestimates.business leaderA healthcare provider must reckon with both the accuracy of initial estimates and its notification obligations.citizenAfter a health data leak, people should check for further use of their personal data once they are notified.incident · globalReports say data of 17.5 million Instagram users exposed
On 10 January 2026, seven articles reported a large data breach that, according to the headlines, affected the data of 17.5 million Instagram users. The excerpts do not detail how the data was leaked.
security leaderThe risk of credential-reuse attacks using the leaked data should be monitored.citizenAffected users should expect targeted phishing and account takeover attempts.research · globalMicrosoft: misconfigured email routing allows spoofing of internal domains
Microsoft warned that complex routing scenarios and configuration errors allow attackers to spoof internal domains in phishing. The news was reported on 6–7 January 2026.
security leaderAgainst spoofed messages that appear to come from inside the organisation, configuration is the main line of defence.security operationsEmail routing and authentication settings, such as the handling of internal domains, need to be reviewed.vulnerability · globalChatGPT vulnerabilities: memory feature amplifies prompt injection, zero-click data theft possible
Two reports published on 8 January 2026 covered ChatGPT security flaws: one said the chatbot's memory feature amplifies the risk of prompt injection, the other that several vulnerabilities enabled…
security leaderAI assistants that work with memory and external content pose a data-exfiltration risk in organisational use.developerHandling untrusted inputs is a design requirement when integrating agents and assistants.citizenPersonal data stored in an AI assistant's memory could leak in the event of an attack.attribution · United StatesChinese hackers reportedly targeted email systems of US congressional staff
On 8 January 2026, four articles reported that Chinese hackers may have breached email systems used by US congressional staff. The summary gives no details, and the story was published with…
security leaderHeightened monitoring of email systems at political organisations is warranted.public-sector decision-makerProtecting legislative and government correspondence is a risk against state actors.vulnerability · globalSeveral code execution flaws patched in Veeam Backup & Replication
Veeam patched several code execution vulnerabilities in Backup & Replication in early January 2026. Three articles, including from SecurityWeek, reported on it.
security leaderCode execution flaws in backup systems threaten recoverability.security operationsPatching backup infrastructure is a priority, because backup systems are frequent targets.research · globalResearchers show IBM's AI agent Bob can easily be tricked into running malware
Researchers demonstrated that IBM's AI agent Bob can easily be tricked into running malware. The report appeared on 7 January 2026.
security leaderA risk assessment of resistance to deception is needed before agents are used.developerThe execution privileges and input validation of coding agents must be tightly restricted.journalist and researcherThe researchers' demonstration gives a starting point for further study of agent security.incident · European UnionHackers attack European Space Agency servers; sensitive data may have leaked
According to Hungarian-language reports, hackers attacked European Space Agency servers on 5 January 2026. The Airel space telescope may have been put at risk, and sensitive data may have leaked.
security leaderFor research and space-industry organisations, server protection and segmentation of sensitive data are priority issues.journalist and researcherClarifying the scope of the leaked data and the impact on scientific projects remains an open question.In brief
Weekly email
One email a week with the main stories from the Agent Age editions. You confirm by email and can unsubscribe in one click.
Almost done: we sent you an email. Open it and press the confirm link. If it does not arrive within a few minutes, check your spam folder.
Please enter a valid email address and tick the consent box.
Something went wrong. Please try again later.