Secure and private AI processing in the cloud poses a formidable new challenge. To support advanced features of Apple Intelligence with larger foundation models, we created Private Cloud Compute (PCC), a groundbreaking cloud intelligence system designed specifically for private AI processing. Built with custom Apple silicon and a hardened operating system, Private Cloud Compute extends the industry-leading security and privacy of Apple devices into the cloud, making sure that personal user data sent to PCC isn’t accessible to anyone other than the user — not even to Apple. We believe Private Cloud Compute is the most advanced security architecture ever deployed for cloud AI compute at scale.
Leaked internal documents have exposed the activities of a Russian state-backed legal defence foundation that European intelligence agencies and analysts say is in fact a Kremlin influence operation active in 48 countries across Europe and around the world.
Internal documents from the Fund for Support and Protection of the Rights of Compatriots Living Abroad (Pravfond) indicate that the foundation finances propaganda websites targeted at Europeans, helped pay for the legal defence of the convicted arms trafficker Viktor Bout and the assassin Vadim Krasikov, and has employed a number of former intelligence officers as the directors of its operations in European countries.
While implementing PHP, the team did not notice the Best-Fit feature of encoding conversion within the Windows operating system. This oversight allows unauthenticated attackers to bypass the previous protection of CVE-2012-1823 by specific character sequences. Arbitrary code can be executed on remote PHP servers through the argument injection attack.
Sophos Managed Detection and Response initiated a threat hunt across all customers after the detection of abuse of a vulnerable legitimate VMware executable (vmnat.exe) to perform dynamic link library (DLL) side-loading on one customer’s network. In a search for similar incidents in telemetry, MDR ultimately uncovered a complex, persistent cyberespionage campaign targeting a high-profile government organization in Southeast Asia. As described in the first part of this report, we identified at least three distinct clusters of intrusion activity present in the organization’s network from at least March 2023 through December 2023.
The three security threat activity clusters—which we designated as Alpha (STAC1248), Bravo (STAC1870), and Charlie (STAC1305) – are assessed with high confidence to operate on behalf of Chinese state interests. In this continuation of our report, we will provide deeper technical analysis of the three activity clusters, including the tactics, techniques, and procedures (TTPs) used in the campaign, aligned to activity clusters where possible. We also provide additional technical details on prior compromises within the same organization that appear to be connected to the campaign.