Latest cybersecurity reporting from selected sources.

AI Tool Used to Discover Bugs, Which Included 2 Maximum Severity VulnerabilitiesResearchers at security firm AISLE said they recently identified 38 vulnerabilities, including two maximum-severity zero-day flaws in OpenEMR, an open-source electronic medical record software platform used by about 100,000 healthcare providers globally. OpenEMR has patched the problems.

Everything is dumb again. This week feels broken in a very familiar way. Old tricks are back. New tools are doing shady crap. Supply chains got hit. Fake help desks worked. Weird research showed how easy some attacks still are

Researchers Detail Prompt Injection, API and Redirect FlawsOasis Security researchers found three bugs in Claude that attackers can chain to steal user chat data without malware or phishing. The "Claudy Day" attack links hidden prompt injection, Anthropic's Files API and an open redirect. Anthropic has fixed the core flaw.

Prompt injection for the win Anthropic has fixed three bugs in its official Git MCP server that researchers say can be chained with other MCP tools to remotely execute malicious code or overwrite files via prompt injection.…

Encryption protects content, not context Mischief-makers can guess the subjects being discussed with LLMs using a side-channel attack, according to Microsoft researchers. They told The Register that models from some providers, including Anthropic, AWS, DeepSeek, and Google, haven't been fixed, putting both personal users and enterprise communications at risk.…

QNAP has fixed seven zero-day vulnerabilities that security researchers exploited to hack QNAP network-attached storage (NAS) devices during the Pwn2Own Ireland 2025 competition. [...]

Plaintext transmissions, fixed MAC addresses, rotating 'unique' IDs, and more, make abuse easy Tile Bluetooth trackers leak identifying data in plain text, giving stalkers an easy way to track victims despite Life360's security promises, a group of Georgia Tech researchers warns.…

More fun with AI agents and their security holes A now-fixed flaw in Salesforce’s Agentforce could have allowed external attackers to steal sensitive customer data via prompt injection, according to security researchers who published a proof-of-concept attack on Thursday. They were aided by an expired trusted domain that they were able to buy for a measly five bucks.…

Google fixed a bug that allowed maliciously crafted Google Calendar invites to remotely take over Gemini agents running on the target's device and leak sensitive user data. [...]

Not a very smart home: crims could hijack smart-home boiler, open and close powered windows and more. Now fixed Black hat A trio of researchers has disclosed a major prompt injection vulnerability in Google's Gemini large language model-powered applications.…

Cybersecurity researchers have disclosed a now-patched, high-severity security flaw in Cursor, a popular artificial intelligence (AI) code editor, that could result in remote code execution

Oracle Cloud Infrastructure Flaw Enabled Malicious File Uploads, Researchers FoundExploring Oracle Cloud Infrastructure, researchers at Tenable found that Oracle's console-based Code Editor tool failed to block arbitrary file uploads, and could be silently exploited via drive-by attacks to install malware. They said Oracle has now fixed the vulnerability.

Hackers Can Use Unverified Email to Log onto SaaS Apps With Entra IDA flaw in a Microsoft single sign-on feature allowing cloud app account takeovers discovered in 2023 never really went away, say researchers - notwithstanding a computing giant claim that it almost immediately fixed the vulnerability known as nOAuth.

Cybersecurity researchers have disclosed details of a now-patched privilege escalation vulnerability in Google Cloud Platform (GCP) Cloud Run that could have allowed a malicious actor to access container images and even inject malicious code

Attacker With Project Access Could Have Retrieved Private Images, Researchers SaidGoogle has fixed Google Cloud Platform vulnerability attackers could have exploited to gain unauthorized access to private container images, due to access restriction shortcomings. Researchers said the flaw highlights how services built atop other services can pose unexpected security risks.

Now-Fixed Flaw Is Big Sleep's First Real-World Bug Find, Say ResearchersGoogle's "highly experimental" artificial intelligence agent Big Sleep has autonomously discovered an exploitable memory flaw in popular open-source database engine SQLite. The researchers detail how the AI agent discovered the now-patched vulnerability.

QNAP has fixed a critical zero-day vulnerability exploited by security researchers on Thursday to hack a TS-464 NAS device during the Pwn2Own Ireland 2024 competition. [...]

Intel works closely with academic researchers on hardware flaws and coordinates efforts with other vendors to roll out fixes for emerging vulnerabilities. That wasn't always the case.

Multiple security vulnerabilities have been disclosed in LG webOS running on its smart televisions that could be exploited to bypass authorization and gain root access on the devices

Google has fixed another zero-day vulnerability in the Chrome browser, which was exploited by security researchers during the Pwn2Own hacking contest last month. [...]