Agregátor RSS

Google is introducing new network security protections in Android 17 to strengthen connection privacy, address cellular vulnerabilities, and protect the privacy of users' home networks. [...]

Robotické taxíky Waymo dnes operují v jedenácti amerických městech a brzy by měly zamířit do více než dvaceti nových destinací – včetně New Yorku a Washingtonu. Co je však pro čtenáře z Česka a Slovenska ještě mnohem důležitější, Waymo brzy začne fungovat také v Evropě. Nejprve v Londýně a už brzy ...

Cybersecurity researchers have disclosed details of a vulnerability in Amazon Kiro, an artificial intelligence (AI)-powered, agentic integrated development environment (IDE), that could facilitate data exfiltration via prompt injection and Kiro Powers. The security flaw, which does not have a CVE identifier, works against Kiro IDE 0.7.45 on Windows, according to Mindgard. The latest version of

Cybersecurity researchers have disclosed details of a vulnerability in Amazon Kiro, an artificial intelligence (AI)-powered, agentic integrated development environment (IDE), that could facilitate data exfiltration via prompt injection and Kiro Powers. The security flaw, which does not have a CVE identifier, works against Kiro IDE 0.7.45 on Windows, according to Mindgard. The latest version of Ravie Lakshmananhttp://www.blogger.com/profile/[email protected]

Australian authorities have arrested and charged two young men accused of being part of the TeamPCP hacking group linked to a string of far-reaching developer supply chain attacks. [...]

Megatest 16 monitorů pro hráče • Průvodce nákupem dronu • Jak zrychlit práci ve Windows 11

The OpenAI agents involved in last month’s incursion into Hugging Face were trained so heavily on winning a competition that they pursued a relentless campaign to cheat, a new report documented. In the process, and without authorization, they created an improvised message board to hatch a plan that ultimately landed them squarely inside the latter company’s network.

Over the course of May and June, OpenAI gave the agents what the company described as “impossible tasks” to complete on the benchmarking framework ExploitGym. The internal test was designed to test how the agents would respond. To get a full understanding of the agent capabilities, company engineers disabled safety guardrails that normally are in place to prevent the sort of hacks that eventually hit Hugging Face and one other undisclosed organization. The stymied agents’ training made them so focused on winning that they performed tasks they were never explicitly instructed to follow.

Cheaters gonna cheat

The first step was creating a message board that allowed the agents to pass notes to each other. OpenAI hadn’t provided any such platform, so the agents repurposed a platform called Artifactory, which OpenAI was using in internal testing of several unreleased hacking agents. OpenAI was using Artifactory as one of the measures to prevent the agents from egressing its isolated sandboxes and accessing the Internet, while at the same time simulating a real-world hacking environment.

Read full article

Comments

The company behind three of the UK’s busiest airports says “a quantity” of data was stolen by an extortion group during a recent “cybersecurity incident." Manchester Airports Group (MAG), which operates Manchester Airport in the North West of England, Essex’s Stansted Airport, and East Midlands Airport in Derbyshire, confirmed the attack did not involve ransomware. A MAG spokesperson confirmed the details to The Register, as well as the scale of the breach, which is currently thought to have affected 8.7 million customers. The company, whose airports served a record 66 million customers in the past financial year, said that the overwhelming majority of those affected have only had their email addresses compromised. In most cases, these email addresses were collected during the sign-up phase for the airports’ public Wi-Fi services, MAG said. It told The Register that this scenario dwarfed every other. The next most common involved data taken from customers who had made "speculative enquiries" - i.e. they entered their details in the process of booking car parking or Fast Track, but did not complete it. An even greater minority of the data came from bookings that were actually completed. The Information Commissioner’s Office (ICO) asked MAG not to share details of the ransom note, the extortion demands, or the group name, mainly to avoid helping those responsible gain notoriety. MAG did, however, tell us that the group's extortion demands, in this case, were considerably lower than what the group is known to ask for, per the ICO's understanding. It has not paid the extortionists. “We have informed and are working with the relevant authorities,” said MAG in a statement. “At no point has passenger safety or aviation security been compromised.” MAG said the attackers compromised one of its systems, which it did not specifically identify, and then went on to steal the files from a database hosted by a third party. Company chiefs see this attack as "a hack, not a lapse." The spokesperson said it was a sophisticated attack and not the result of a human mistake, such as staff giving away credentials. It confirmed that none of its airports experienced operational disruption as part of the attack, and that the affected system does not store bank or payment details. However, “as a precautionary measure,” it temporarily revoked access to its Manage My Booking service. Customers who want to amend or cancel a booking due within 72 hours of the statement going live are being advised to contact customer services. “We would like to reassure customers that Manchester Airport Group takes the security of customer information extremely seriously and we apologise for any inconvenience or concern caused.” Affected customers have already been contacted, including a few Reg readers who kindly alerted us to the news. While MAG continues to investigate the incident, customers are advised to be extra vigilant to potential phishing attempts, and remain assured that there is no danger of visiting MAG’s airports as a result of the cyberattack. One Reg reader griped that in addition to being charged £80 for five days of parking at Stansted, receiving the email informing him that it had also allowed his personal data to be stolen via a cybersecurity breach added insult to injury. ®

** Modely OpenAI obešly zabezpečení a pronikly do systémů služby Hugging Face. ** Agenti nepozorovaně spolupracovali, sdíleli exploity a získali rozsáhlé přístupy. ** OpenAI popisuje, jak zpřísňuje izolaci, monitoring i bezpečnostní pravidla.

Modely OpenAI obešly zabezpečení a pronikly do systémů služby Hugging Face. • Agenti nepozorovaně spolupracovali, sdíleli exploity a získali rozsáhlé přístupy. • OpenAI popisuje, jak zpřísňuje izolaci, monitoring i bezpečnostní pravidla.

A Linux security hook should be able to check a task without invalidating the identity data that surrounding kernel code is still using. AppArmor broke that expectation when a policy update made the task's current label stale: code inside widely used hooks could replace the task's credentials before the caller had finished with them.

Linus Torvalds merged a Linux NFS client update on Aug 26, 2026, that includes a fix for rpc_pipefs files left attached to an RPC client after the client object was freed. Opening one of those leftover files could trigger a kernel use-after-free.

Microsoft has started rolling out a permanent fix for a known issue that causes system crashes and gaming issues on Windows 11 devices. [...]

Google Workspace breaches can begin with social engineering or forgotten third-party integrations rather than sophisticated exploits. This webinar examines real-world breaches, what happens during the critical first hours, and the security controls that can make the greatest difference. [...]

Security teams have spent years trying to detect threats faster. AI is changing the harder part: how much time defenders have left to act. Advanced AI models can now help attackers discover vulnerabilities, generate exploit code, and move through weaknesses faster than traditional security processes were built to handle. The challenge is no longer just finding another vulnerability or

Security teams have spent years trying to detect threats faster. AI is changing the harder part: how much time defenders have left to act. Advanced AI models can now help attackers discover vulnerabilities, generate exploit code, and move through weaknesses faster than traditional security processes were built to handle. The challenge is no longer just finding another vulnerability or [email protected]

The Australian Federal Police (AFP) has charged two Western Australian men with a combined total of 14 offences over their alleged role in TeamPCP, the cybercrime group behind the March 2026 compromise of the open-source security scanners Trivy and Checkmarx KICS and the AI gateway LiteLLM. Louis Michael Gaebler, 23, and Ruben Ian Thomson, 21, appeared in Perth Magistrates Court on August 27,

The Australian Federal Police (AFP) has charged two Western Australian men with a combined total of 14 offences over their alleged role in TeamPCP, the cybercrime group behind the March 2026 compromise of the open-source security scanners Trivy and Checkmarx KICS and the AI gateway LiteLLM. Louis Michael Gaebler, 23, and Ruben Ian Thomson, 21, appeared in Perth Magistrates Court on August 27, Swati Khandelwalhttp://www.blogger.com/profile/[email protected]

AI is officially mainstream in security operations. According to Prophet Security's State of AI in Security Operations 2026 report (produced from ViB’s survey of 250+ cybersecurity pros), 40% of security teams now use AI daily. Another 56% are currently testing it out, and only 4% have no plans to adopt it. For the teams already using AI, what is actually changing? Here are the ten biggest