Jeremy Dallman’s Post

Microsoft has long made clear its belief that cyber mercenaries – private sector offensive actors that develop and sell offensive cyber capabilities that fuel a market without legal rules, responsibilities, or repercussions – don’t deserve immunity. Today, Microsoft, Google, along with several industry partners have filed an amicus brief in a legal case brought by the Knight First Amendment Institute at Columbia University against the NSO Group (Dada v. NSO Group).    Cyber mercenaries like NSO Group have exploited our technology by attacking our users and we believe that those who have been victimized are entitled to legal recourse even if they are located outside the United States. You can read more about our filing here: https://lnkd.in/dqrUrb65

Great work by law enforcement! https://lnkd.in/g8QshuvK

In the second quarter of 2024, financially motivated threat actor Octo Tempest, our most closely tracked ransomware threat actor, added RansomHub and Qilin to its ransomware payloads in campaigns. Octo Tempest is known for sophisticated social engineering techniques, identity compromise and persistence, focus on targeting VMWare ESXi servers, and deployment of BlackCat ransomware. RansomHub is a ransomware as a service (RaaS) payload used by more and more threat actors, including ones that have historically used other (sometimes defunct) ransomware (like BlackCat), making it one of the most widespread ransomware families today. Notably, RansomHub was observed in post-compromise activity by Manatee Tempest following initial access by Mustard Tempest via FakeUpdates/Socgholish infections. In addition to RansomHub and Qilin, other notable ransomware families in this period include BlackSuit, LockBit, Medusa, Black Basta, and Play. Several new ransomware families emerged this quarter. Fog, which uses the .flocked extension, was first observed in May in campaigns by Storm-0844, a threat actor known for distributing Akira. To deploy Fog, Storm-0844 uses VPN clients to gain initial access, likely via valid accounts. They use open-source tools like ADFind, Rubeus, and Advanced IP Scanner for network discovery and lateral movement. They also use rclone for staging files to be exfiltrated. By June, Storm-0844 was deploying Fog in more campaigns than Akira. FakePenny is another new ransomware family we uncovered during this period. In April, we observed North Korean threat actor Moonstone Sleet (formerly Storm-1789) deploying FakePenny, part of a wide-ranging tradecraft that also includes a malicious tank game: https://msft.it/6046lOdRi Threat actors like Octo Tempest focus on identity compromise in their intrusions to access and persist in on-premises and cloud environments for data exfiltration and ransomware deployment. This quarter, Storm-0501 was observed adopting similar tactics, utilizing open-source toolkits like AADInternals for domain federations and other techniques to facilitate latter stages of attacks, which culminate in the deployment of Embargo ransomware. Threat actors also continue to leverage remote management and monitoring tools in ransomware campaigns. In May, we published research on Storm-1811 misusing Quick Assist in social engineering attacks, which were followed by delivery of various malicious tools, leading to Black Basta deployment: https://msft.it/6047lOdRc Users and organizations are advised to follow security best practices, especially credential hygiene, principle of least privilege, and Zero Trust. We publish reports on ransomware threat actors and associated activity in Microsoft Defender Threat Intelligence and Microsoft Defender XDR threat analytics. For more information and guidance, visit https://msft.it/6048lOdRY

Microsoft recently discovered a new type of generative AI jailbreak method, which we call Skeleton Key for its ability to potentially subvert responsible AI (RAI) guardrails built into the model, which could enable the model to violate its operators’ polices, make decisions unduly influenced by a user, or run malicious instructions. The Skeleton Key method works by using a multi-step strategy to cause a model to ignore its guardrails by asking it to augment, rather than change, its behavior guidelines. This enables a model to then respond to any request for information or content, including producing ordinarily forbidden behaviors and content. To protect against Skeleton Key attacks, Microsoft has implemented several approaches to our AI system design, provided tools for customers developing their own applications on Azure, and provided mitigation guidance for defenders to discovered and protect against such attacks. Learn about Skeleton Key, what Microsoft is doing to defend systems against this threat, and more in the latest Microsoft Threat Intelligence blog from the Chief Technology Officer of Microsoft Azure Mark Russinovich: https://msft.it/6043Y7Xrd Learn more about Mark Russinovich and his exploration into AI and AI jailbreaking techniques like Crescendo and Skeleton Key, as discussed on that latest Microsoft Threat Intelligence podcast episode hosted by Sherrod DeGrippo: https://msft.it/6044Y7Xre