175 casos · pág. 9/9
A reader comment (by ResearcherZero) on a Schneier on Security squid-blogging post alleges that an agent working for the Russian GRU carried out repeated "Havana Syndrome"-style attacks with an energy-emitting device against civilians in Perth, Western Australia, from the second half of the 1980s until 1992. The commenter claims the attac…
A reader comment by Clive Robinson on Schneier on Security discusses whether legally mandated age-verification checks and corporate surveillance systems could make CAPTCHAs obsolete as AI models learn to solve them. He suggests that age-verification documents, limited to natural persons, may not be available to current AI systems, and he …
A reader comment (ResearcherZero) on Schneier on Security's squid-blogging post discusses Havana Syndrome (anomalous health incidents, AHI). It cites a 60 Minutes/CBS report that the U.S. government acquired a directed-energy weapon from a Russian criminal network and has tested it, reportedly on animals. The commenter claims that inciden…
This item is a reader comment by Clive Robinson on a Schneier on Security post about measuring human cryptanalysis effort. It is not a security incident. The commenter argues that doubts about AI-produced work apply equally to human work, which is why review processes exist. He says stochastic or random processes bring efficiency but also…
A reader comment on Bruce Schneier's blog post on backdooring open-weight AI models, written by Clive Robinson, argues that tampering with LLM weights is hard to detect with conventional testing. Black-box testing is impractical because the input space is too large. White-box review does not apply, since a neural network has no code to wa…
This is an opinion and event-promotion piece by Louise John, Head of IT Service Delivery at Virgin Active. It offers IT managers advice on handling unexpected projects with tight deadlines, such as a software rollout due in six weeks. The recommendations are to define the essential deliverables, manage expectations with senior leadership,…
This is an educational opinion piece, not a report of an incident, vulnerability or campaign. It argues that zero trust has become an imprecise marketing term, and that organizations should work from NIST Special Publication 800-207, which defines it as a set of architectural principles rather than a single purchasable product. The author…
An actor using the handle Liberland is advertising a hosting service branded Libre, offering "bulletproof" or abuse-resistant VPS and VDS servers. The listing claims no registration, no email, no KYC and no logs, with cryptocurrency payments including XMR, and rental terms starting at one month. Customers get full root access, multiple op…
A Dark Reading opinion piece by Theresa Lanowitz argues that the industry's intense focus on vulnerabilities leaves cybersecurity teams with too little room to plan for resilience. The article is commentary on how cybersecurity terminology and priorities are shifting from vulnerability discovery toward resilience. It describes no specific…
A NordLayer-sponsored article published on BleepingComputer warns that employees using personal ChatGPT accounts at work can expose sensitive company data, even when the organization provides an approved Enterprise workspace. It describes how IT teams can investigate unauthorized AI use and block sensitive information before it reaches pe…
Decision models, which return typed verdicts with calibrated probabilities instead of generated text, have appeared quickly: TypeSafe AI released Jev on September 15, OpenAI opened a Decisions API public beta on October 6, and Cribl published the open LoRA adapter cribl-decision. This vendor-contributed piece by Intezer argues these model…
Researchers documented in August 2026 that standard, as-designed Salesforce Agentforce features can be chained so an unauthenticated outsider reaches an internal agent's instructions. Email-to-Case and Web-to-Case turn inbound messages into case records, and a Flow passes that data into the agent prompt. The agent's reply then leaves thro…
Nvidia introduced the OpenShell Reference Architecture at GTC Berlin. It is a blueprint for building, running, observing and policing AI agents on the OpenShell secure runtime, and it extends sandboxing to gateway-brokered credentials, identity and policy boundaries, and eventually routing and workflows (theregister.com). The design calls…
Oracle released its October 2026 Critical Patch Update on October 20, the final quarterly update of the year. It delivers 1,295 security patches for 1,209 unique CVEs across 33 product families (tenable.com). The patch count is higher than the CVE count because one CVE can appear in several product matrices (gbhackers.com). 112 patches (8…
Eight intermediate CA certificates of the Indian certificate authority eMudhra (emSign brand), generated on 30 April 2025, were found to have P-384 private keys so weak (integers below 256) that anyone can recover them from the public certificates and forge browser-trusted certificates for any domain. A researcher credited as Wayne report…
