Sergiy Galyonkin has added a photo to the pool:

The vulnpocalypse is upon us, dear reader. Microsoft delivered a record number of patches to address 974 CVEs in its own products this month, including two bugs that Redmond says are already under exploitation. September's record-breaking collection of security updates come after Microsoft served up 421 fixes in August, and 622 in July. We've seen the new normal and we are not impressed. Thanks, but no thanks, AI. In addition to Microsoft’s massive patch drop, Adobe on Tuesday issued 10 bulletins addressing 172 CVEs, including a max-severity vulnerability exploited as a zero day in Magento and its successor product Adobe Commerce. Adobe on Monday shipped a hotfix for this one, tracked as CVE-2026-75650 and named StyleSmuggler, that gives unauthenticated attackers remote code execution. StyleSmuggler If your organization has any type of online shop, prioritize this one first as it’s already being abused to compromise stores, according to e-commerce security shop Sansec. Sansec discovered StyleSmuggler, and reports that attacks started on September 4. Every version of Magento and Adobe Commerce, from 2.4.4 up to and including 2.4.9, has the flaw. The bug allows attackers to inject malicious PHP code inside Magento templates using the “styles” properties to evade safety detections. In confirmed attacks, the payload then installs a backdoor that connects to a command-and-control server and waits for instructions. “So far, we have no indication that the backdoor has been weaponized,” the Sansec Forensics Team wrote. Don’t wait to find out on this one. Put it at the top of your mitigation list. Microsoft's 974 CVEs On to Microsoft’s record-breaking 974 CVEs, which according to Tenable is not many fewer than the 1,130 CVEs Redmond issued in 2025. Two are already being exploited as zero-days. First up: CVE-2026-85880, a privilege escalation bug in Windows Advanced Local Procedure Call (ALPC). Successful exploitation can result in the attacker gaining SYSTEM privileges. “An attacker who can execute code in a low-privilege AppContainer could exploit this vulnerability locally to escape the sandbox and elevate privileges on the affected system,” Redmond warned. “No additional user interaction is required.” No word yet on who is exploiting this bug, and to what end. The US Cybersecurity and Infrastructure Security Agency on Tuesday added CVE-2026-85880 plus a second Microsoft security hole (and the Adobe Commerce and Magento zero-day) to its Known Exploited Vulnerabilities Catalog, and set a September 22 deadline for federal agencies to fix both new Microsoft bugs and a September 11 deadline to patch the Adobe flaw. The second Microsoft bug found and exploited as a zero-day is CVE-2026-81963, another privilege escalation vulnerability. This one affects the Windows Update Stack. We also have very little detail about this flaw, other than it also allows attackers to gain SYSTEM-level access. “More likely is that this bug is being combined with a code execution bug to spread malware or ransomware,” opined Zero Day Initiative’s Dustin Childs, who advised users to “Patch this one quickly.” While those are the only two (so far) under active exploitation, Childs rated CVE-2026-55007, one of nine Exchange Server flaws disclosed this month, as “the most important” patch for the messaging server. It allows a remote, unauthenticated attacker to execute code on a vulnerable Exchange server by sending an email with a malicious Visio attachment. No user interaction is required, and the code executes when the server processes the attachment during content indexing. Redmond says it’s “difficult to reliably trigger,” but as Childs points out: “The attacker only needs to get it right once. Schedule your downtime and update your Exchange servers with haste.” Childs also said he counts 20 patches for wormable bugs, so be sure to read his full Patch Tuesday review for those. “While some might be more exploitable than others, having 20 of them in a single release is something else.” The missing CVE While Redmond addressed nearly 1,000 security holes this month alone, it’s also worth pointing out one that isn’t this month’s Patch Tuesday roundup: CVE-2026-85046. Google patched this bug in Chrome on September 3, and at the time warned that it “is aware that an exploit for CVE-2026-85046 exists in the wild.” The high-severity, type confusion flaw exists in the V8 JavaScript engine used in both Google’s Chrome and Microsoft’s Edge browsers. And yet Microsoft still hasn’t published a security advisory for CVE-2026-85046. “If you’re patched, you are protected, but if you rely on advisories to know which vulns exist, you could miss this zero-day vulnerability altogether,” Adam Barnett, lead software engineer at Rapid7, told The Register. “A patch without an advisory is perhaps marginally better than an advisory without a patch, but keeping track of exposures without reliable advisory materials is not straightforward,” Barnett said. “Chrome patched 11 other vulnerabilities at the same time as CVE-2026-85046, but it’s not yet clear if those are patched in Edge. Until Microsoft sets the record straight, the only safe assumption is that these vulnerabilities (e.g. CVE-2026-85045) remain unpatched in Edge.” ®

The NSA, CISA, and FBI are accusing (PDF) several Chinese AI companies of carrying out "industrial-scale" distillation campaigns against leading U.S. models such as ChatGPT, Claude, Gemini, and Grok. Since at least 2024, the companies have allegedly routed millions of requests across accounts, APIs, proxies, cloud providers, and third-party aggregators to extract capabilities for their own models. "China-based artificial intelligence companies are conducting systematic extraction of proprietary functionalities and capabilities of U.S. AI companies' models through industrial-scale knowledge distillation campaigns that form the core -- not merely a supplement -- of their AI development strategy," the agencies wrote. CyberScoop reports: DeepSeek, for example, distilled frontier U.S. models to generate synthetic training data for its R1 and R3 models, including four different versions of Claude, two versions of Gemini, five versions of ChatGPT and Grok 4. Those models helped train DeepSeek's capabilities in areas like agentic functioning, question and answer optimization, creative and occupational writing and others.

Another Chinese company, Moonshot AI, allegedly distilled 18 different U.S. models -- including Fable 5, Anthropic's current, most advanced commercially available model -- to train its Kimi-K2 and Kimi K3 models. The company used millions of queries meant to extract enhanced capabilities in areas like agentic reasoning, coding and data analysis, computer vision, larger logical frameworks, visual processing and others.

Chinese AI companies manage a sophisticated set of tools and systems that route requests and prompts through multiple pathways to avoid detection. The advisory lists common tactics observed by Chinese companies, including spreading requests across different accounts, models and platforms, using native APIs, remote cloud providers, and third-party aggregators to obfuscate user metadata, and leveraging proxies and gray tech markets to get around geographic restrictions, terms of use and safeguards built into frontier models.

Read more of this story at Slashdot.

The affluent city-state has justified high-end salaries for its political leaders, arguing they attract talent and ensure good governance

Singapore’s prime minister has been given a $1m pay rise as the affluent city-state announced a dramatic one-off salary boost for government ministers.

The pay increases add more than 60% to what are already the world’s highest ministerial salaries.

Continue reading...