Researchers tested 31 million patterns to disrupt surveillance AI, with promising results but significant gaps between simulation and real-world use. The Kansas City-based cybersecurity researcher Bill Swearingen spent the past year doing something that sounds almost too simple to work: printing patterns, watching cameras fail to detect them, and repeating. TechCrunch reports that after roughly 31 million tests, he can now generate patterns on demand that block license plate readers and surveillance cameras from recognizing whatever the pattern covers, whether that’s a person or a vehicle. The project is called noRecognition, and the core idea isn’t stealth in the traditional sense. The camera still records everything just fine. What breaks is the detection layer sitting on top of the footage, the software that flags license plates, tracks faces, or spots “activity of interest” across thousands of hours of video. Swearingen’s patterns don’t hide you from the lens; they make the algorithm looking through that lens shrug and move on. Swearingen, co-founder of the SecKC meetup, said his project started for personal reasons. He became concerned about the growing number of surveillance cameras in his town and the possibility of being tracked while attending a protest. What started as a simple experiment later became a reinforcement learning system. He taught the model to create patterns, learn from failures and keep improving. Over time, it learned how to avoid detection by several camera systems. Every time a pattern failed and got detected, the system adjusted and tried again, eventually learning to defeat multiple detection algorithms simultaneously rather than just one at a time. The research dashboard behind the project, published at sandbox.norecognition.org, goes considerably deeper into the numbers than the headline claim suggests, and it’s refreshingly upfront about what’s proven versus what isn’t. The team states its overall objective plainly as “one pattern that defeats every detector,” and by their own account that goal remains only partially met. Their strongest validated result against a detector extracted directly from a real deployed surveillance camera sits at 61.7% non-detection across held-out test subjects, a solid number, but nowhere near total, and still a digital simulation rather than a real-world fabric test. That distinction matters more than it might seem. Most of the dashboard’s headline figures are explicitly labeled as digital, simulated results, meaning the pattern was tested against a virtual camera and printed ink model rather than an actual garment photographed by an actual camera in the field. The gap between “works in simulation” and “works when Donut Media wraps a real 2009 Toyota Yaris in it,” which is the physical test Swearingen ran live at DEF CON, is exactly the gap this kind of research has to close before anyone should treat it as a reliable, everyday privacy tool. “On Friday at the Def Con cybersecurity conference in Las Vegas, Swearingen ran his first real-world test. With help from Donut Media, the test involved covering a 2009 Toyota Yaris with one of Swearingen’s newest patterns to see if the car would be invisible to detection by a Flock camera.” reports TechCrunch. “We proved it was effective,” said Swearingen, though the wheels were a challenge. The video of the demo will be out in the next few weeks, said Donut Media.” That DEF CON demo is where things got concrete. Swearingen covered a car in one of his newest patterns and tested it against a Flock Safety camera, the kind widely deployed for automated license plate reading across the US. He said the test proved effective, though the vehicle’s wheels turned out to be a persistent weak point, curved surfaces apparently don’t cooperate with flat printed patterns the way a car door does. Source Tech Crunch – A photo of a 2009 Toyota Yaris at the Def Con conference in Las Vegas, covered in a pattern made by Bill Swearingen, as part of a test to see if it can defeat surveillance camera detection. Image Credits:Bill Swearingen / Donut Media Swearingen is not publishing his best patterns because he does not want camera makers to easily find and block them. Instead, he is using crowdfunding to develop and sell printed products such as T-shirts and hoodies, with vehicle wraps possibly coming later. It is still unclear whether the project will become a practical privacy tool for everyday users or remain mainly a DEF CON demonstration. Its real effectiveness will depend on how well the patterns work on real clothing, in different weather and camera conditions. Follow me on Twitter: @securityaffairs and Facebook and Mastodon Pierluigi Paganini (SecurityAffairs – hacking, Surveillance camera)
Security controls can block a familiar attack method while missing quieter ways to achieve the same objective. Picus Security's Blue Report 2026 shows how prevention rates can vary dramatically by technique and why behavioral testing is needed to uncover those gaps.
Microsoft says some users are experiencing issues searching in Microsoft 365 apps, including Outlook on the web, Outlook desktop, SharePoint Online, and OneDrive. According to an incident report seen by BleepingComputer and tracked under MO1456424 in the Microsoft 365 Admin Center, the root cause is what Microsoft describes as a recent deployment that causes resource utilization problems. "Impact is specific to some users served through the affected infrastructure who are attempting to search for content in SharePoint Online, OneDrive, Outlook on the web, or Outlook desktop," Microsoft said. "Our investigation identified that a recent deployment introduced a resource utilization inefficiency issue, leading to impact." Microsoft says it has already developed a fix and deployed it to reduce resource pressure and restore service for all affected Microsoft 365 users.
GitLab patched a critical GraphQL flaw that let unauthenticated attackers remotely modify or delete public projects on self-managed servers. GitLab pushed out an emergency patch this week to address a critical flaw, tracked as CVE-2026-19478 (CVSS score of 9.4), that could let an attacker with zero credentials remotely modify or delete public projects and user data. “GitLab has remediated an issue that under certain conditions could allow an unauthenticated user to remotely modify or delete public projects and user data via a GraphQL directive.” reads the advisory. GitLab issued an emergency patch on August 17, five days after its regular update. The vulnerability impacts only self-managed installations, users should upgrade to versions 19.2.4, 19.1.6, 19.0.8, and 18.11.11. There’s a gap worth flagging for anyone still sitting on an older release. The available patches don’t cover the 18.2 through 18.10 branches, even though those versions technically fall inside the affected range. If you’re running anything in that window, staying put isn’t really an option; you’ll need to upgrade to a patched branch entirely rather than waiting for a fix that isn’t coming for your current one. hiimguardian reported the flaw through the company HackerOne bug bounty program. A second, less severe issue shipped in the same release. CVE-2026-19650 (CVSS score of 7.1), involves a cross-site request forgery weakness in how GitLab’s GraphQL handles multiplex queries, letting an unauthenticated attacker trigger mutations through GET requests due to improper validation. Unlike the critical flaw, this one needs a victim to actually interact with something, so it’s a real risk but a considerably narrower one. There’s no evidence either bug has been exploited in the wild yet, and no public proof-of-concept code has surfaced as of publication. That’s the good news. The less comforting part is GitLab’s own disclosure policy: full technical details typically go public on the company’s issue tracker 90 days after the patching release, which puts a working understanding of exactly how this bug functions squarely in mid-November, plenty of time for someone motivated enough to reverse-engineer the patch diff themselves before then. This is not the first serious GitLab issue in recent months. Last month, researchers released working exploit code for another remote-code-execution flaw affecting self-managed servers. Organizations running GitLab on their own infrastructure should therefore treat this patch as a priority, especially given the critical severity and the fact that the vulnerability requires no authentication or user interaction. Follow me on Twitter: @securityaffairs and Facebook and Mastodon Pierluigi Paganini (SecurityAffairs – hacking, GitLab)
Microsoft announced that it removed the Windows Management Instrumentation Command-line (WMIC) tool from Windows 11 24H2 and 25H2, as well as from Windows 11 beta builds released this week. WMIC is a legacy built-in Windows command-line utility that helps interact with the Windows Management Instrumentation (WMI) system using text commands. This move is part of a process announced in September, when the company said that WMIC will be removed after upgrading to Windows 11 25H2 and later. Microsoft deprecated WMIC in Windows Server 2012 (in 2016) and Windows 10 21H1 (in 2021), and it converted it into a Feature on Demand (FoD) starting with Windows 11 22H2 (in 2022), and announced in January 2024 that it would be removed altogether after first disabling it by default. "Windows Management Instrumentation Command-line (WMIC) has been removed in this release. This change is part of the ongoing deprecation and removal of WMIC from Windows," Microsoft noted on Monday.
Varonis Threat Labs has disclosed three vulnerabilities in Microsoft Copilot Personal that it said could allow a single click on a crafted link to silently pull data from connected apps and other information available to the victim's Copilot session. The flaws, which the researchers collectively named CoSnitch, turn in part on an undocumented URL parameter that the assistant itself surfaced
Security researchers at Anthropic and Switzerland's EPFL have demonstrated that self-propagating payloads can spread from one artificial intelligence (AI) agent to the next through the editable system prompt files that autonomous agent harnesses use to carry state between sessions. The work, released as a preprint on August 10, 2026, tests the technique in a simulated six-agent coding
Cybersecurity researchers have disclosed details of a previously undocumented Python implant framework dubbed TWINLOOT. "TWINLOOT is a modular, PyArmor-hardened Python implant designed to operate its entire command-and-control infrastructure inside trusted Microsoft services," Ontinue said in a technical report shared with The Hacker News. "Tasking flows through SharePoint Online file
Cybersecurity researchers have flagged a new typosquatting campaign targeting RubyGems users with a Windows-based information stealer. OpenSourceMalware, which discovered the activity on August 15, 2026, is tracking the threat under the moniker StubMaker. The complete list of packages published as part of the campaign is below - ubnuler ubnlder ri18nr reaker rakier orakw joxn
SafePal has disclosed that an authorization flaw in an order-tracking plug-in exposed the names, email addresses, shipping addresses, phone numbers, and purchase details of approximately 39,798 customers. The hardware wallet maker said all affected customers were notified individually by email on August 16 from security@safepal.com, with the subject line "[Important] Your SafePal Order
Confirm this action.
Leaving now will discard your changes.