Windows 11 Update KB5124010 Triggers Crashes in Battlefield 6 and Other Titles
An optional Windows 11 update identified as KB5124010 is creating serious disruptions for gamers, particularly those playing Battlefield 6 and Battlefield 2042. The patch, released in early September, was first offered to participants in the Windows Insider program for versions 24H2 and 25H2 before being distributed more widely as a non-security update.
Users on Reddit and Steam report that affected games close abruptly after 15-30 seconds or within a few minutes of starting, often without displaying any error code. Standard troubleshooting steps such as reinstalling the game, verifying DirectX installations, and attempting to repair anti-cheat components have not resolved the crashes.
Removing KB5124010 has restored normal operation in numerous reported cases, pointing to a likely conflict between the update and anti-cheat systems. Battlefield 6 and Battlefield 2042 rely on Electronic Arts' Javelin anti-cheat, while similar issues have appeared in titles protected by Easy Anti-Cheat, including Wardogs which displays error code WD-L020-95bc94d86a0d.
The problems occur across varied hardware setups, making it difficult to attribute them to specific graphics cards or processors. Microsoft has not yet confirmed any connection to anti-cheat software, and all conclusions are based on user reports. Because the update is optional, players are recommended to avoid installing it until a resolution is available.
Related articles
NICT Warns of A2B Attacks Where Reverse-Engineered Android Apps Expose Vulnerable Backend Servers
Japan's National Institute of Information and Communications Technology (NICT) has issued an alert on A2B attacks, in which publicly available Android apps are reverse-engineered to identify and target their backend servers. The technique, called App-to-Backend Attack, allows threat actors to discover internal server endpoints even when organizations do not publish connection details. Once identified, attackers exploit vulnerabilities, weak authentication, or misconfigured APIs on those servers, leading to unauthorized access and data theft. NICT notes that multiple recent membership-service breaches in Japan stem from this vector rather than flaws in the mobile apps themselves. The institute urges developers and operators to harden backend infrastructure, conduct thorough compromise assessments, and apply proper access controls. Related incidents include unauthorized access to Lawson ID services, Yakiniku King app member data, and Japan Post's postal app.
From HFT to Infosec: How AI Accelerates the Developer-Hacker Arms Race
An experienced security researcher draws a parallel between high-frequency trading evolution and modern vulnerability research, showing how artificial intelligence compresses the time between patch release and working exploit. The author describes the classic workflow where developer Rob fixes buffer overflows in an FTP server while researcher Dave reverse-engineers the diff to locate the exact change. With AI assistance, both sides now generate patches or exploits in minutes rather than days or weeks. The piece highlights that publishing a patch effectively discloses the location of the flaw, allowing automated tools to produce reliable exploits almost instantly. Historical examples such as wu-ftpd CVEs and the EternalBlue/MS17-010 worm illustrate how quickly mass exploitation follows disclosure. The author warns that open-source projects face an existential dilemma: withhold patches and leave users exposed, or release them and trigger immediate automated scanning. Recommendations include mandatory back-porting of fixes and wider use of runtime protections, yet the overall outlook remains pessimistic about the coming wave of fully automated attacks.
Ghost Assets in Vulnerability Management: How Identification, Merging and Asset History Eliminate Anomalies in MaxPatrol VM
Positive Technologies experts explain how infrastructure changes create duplicate, merged and phantom assets that distort vulnerability management. The article details three core mechanisms inside MaxPatrol VM: unique asset identification across scan sources, non-destructive merging of new and existing data, and full lifecycle history with configurable aging policies. It describes how the system distinguishes assets using prioritized keys such as system ID, FQDN, MAC address and VM ID, then applies recursive merging rules that respect multiple data sources. Policies allow teams to set freshness and obsolescence thresholds, automatically hiding assets after 90 days while preserving historical snapshots for PDQL queries. The piece also covers practical cloning and partial-scan scenarios that can produce misleading duplicates or lingering collections.
GreyNoise Detects Active Exploitation Attempts of CVE-2021-36260 Against Hikvision Cameras
GreyNoise recorded exploitation attempts targeting CVE-2021-36260 in Hikvision cameras and recorders between September 21 and October 1. The vulnerability carries a CVSS score of 9.8 and allows unauthenticated remote command execution through insufficient input validation in the embedded web server. Attack traffic originated from four IP addresses, three routed through a commercial VPN in Lithuania and one from a Ukrainian residential network, with all targets located in Ukraine. Attackers leveraged a publicly available Nuclei template to probe devices, though GreyNoise confirmed only scanning activity and not successful compromises. The five-year-old flaw remains actively scanned despite an available firmware patch from Hikvision and a CISA advisory recommending immediate updates. Password changes provide no protection because the attack requires no authentication. Defenders are advised to apply firmware updates, remove devices from public internet exposure, and segment video surveillance networks from critical infrastructure.