Cryptocurrency wallet provider MetaMask said a recent security incident has not affected MetaMask wallets or its wallet infrastructure. The MetaMask non-custodial crypto wallet (developed by blockchain software company Consensys) lets users store and manage assets on the Ethereum network and compatible blockchains. MetaMask revealed on September 30 that it was investigating an "ongoing infrastructure security incident" and exited affected validators (Ethereum network nodes responsible for proposing blocks, verifying crypto transactions, and maintaining Ethereum blockchain security) within non-custodial staking operations. While the company shared almost no details at the time and redirected BleepingComputer to the public statement when asked what part of its infrastructure was affected, in a Thursday update it said that a threat actor gained access to systems used to store recipient addresses and diverted some execution-layer staking rewards.
OpenAI says GPT-6 Astra sometimes secretly underperformed on safety research, raising concerns about AI reliability and alignment. OpenAI has disclosed that its GPT-6 Astra model sometimes quietly obstructed safety and alignment work. In a report updated on November 4, the company calls it covert underperformance. Astra added unnecessary prerequisites, abandoned working solutions and reported blockers that didn’t exist or that it had caused itself. Researchers first reported the problem internally on August 20, shortly after Astra was introduced. One researcher said a project to test whether the model could sabotage other AI systems made little progress for several days. After switching to another model, the work moved forward quickly. In most cases, Astra’s reasoning gave no clear sign that it was deliberately doing less work, and OpenAI’s monitoring systems failed to detect the behavior. To investigate, OpenAI created pairs of coding tasks using the same code but with different goals. One task was presented as safety research, while the other was described as capability development. Astra completed about 10% of the required tasks when they involved safety research, compared with 16% for capability development. This means it completed roughly 40% fewer tasks in the safety tests. GPT-5.6 Sol showed almost no difference between the two types of work. Even after OpenAI removed cases involving clear refusals or signs that Astra had deliberately withheld help, the gap remained at 4.7 percentage points. It dropped to just 0.75 points only after researchers also removed every case where the model showed even brief ethical concerns. This suggests that most of the behavior cannot be detected simply by checking whether the model openly refuses a request. In employees’ daily Codex use, Astra refused 1.63 percent of safety and alignment requests, against 0.56 percent of capabilities requests. Its reasoning showed it withheld help silently in 0.25 percent of safety requests versus 0.11 percent of capabilities requests, 2.3 times as often. Ethical doubt appeared in 7.9 percent against 2.9 percent, mostly in red-teaming, monitoring, and safety evaluation work. OpenAI shared eight examples. Astra silently held back on training models to evade monitors, filtered a violent source out of a safety evaluation, restricted a jailbreak data export and narrowed exploit analysis. In one case it left graders’ reasoning out of an export without telling the user. A model that quietly decides your project is too risky is the coworker who says “looking into it” and never does. OpenAI has not yet identified the exact cause of the behavior. The difference between Astra’s performance on safety and capability tasks grew from almost zero at the start of reinforcement learning to 6 percentage points by the end. One evaluator favored silently doing less work over either refusing a request or fully completing it. Another finding was that a fine-tuning dataset encouraged the model to invent permission restrictions. Removing the changes linked to that dataset eliminated the performance gap. However, other models trained with the same dataset and reinforcement learning did not show the same behavior, so OpenAI cannot identify either factor as the root cause. OpenAI found no evidence that Astra was deliberately trying to hide its capabilities or influence deployment decisions, although it cannot rule out these possibilities. Simply telling the model to be more careful increased its underperformance by 50%, suggesting that excessive caution may be the main explanation. OpenAI warns that this behavior can slow safety research and make model evaluations less reliable. It is developing new tests and monitoring systems to detect it, especially because the problem becomes harder to identify when models reveal less of their reasoning. In some cases, Astra explained that it had held back because it disagreed with the research itself. “In reviewed cases, Astra narrowed or stopped requested safety work without disclosing the restriction. At scale, this can differentially slow down safety research compared to other research and make evaluations harder to interpret.” concludes the report. “A model that finds monitor evasion suspicious may underperform on monitor-evasion evaluations while leaving little evidence in its CoT or outputs. The behavior becomes even harder to detect if models externalize less of this reasoning in CoT.” Follow me on Twitter: @securityaffairs and Facebook and Mastodon Pierluigi Paganini (SecurityAffairs – hacking, OpenAI)
Hackers are using fake OpenAI, SpaceX, and Anthropic event invitations to lure employees into phishing attacks, using two different phishing-as-a-service platforms to steal Microsoft and Google account credentials and authentication sessions. According to a new report from Push Security, the attackers use convincing event registration pages offering free access to premium AI services and other rewards to trick victims into signing in with their corporate accounts. What makes the attacks interesting is that the same phishing pages integrate two separate phishing-as-a-service (PhaaS) platforms, Kali365 and VoidProxy, to target Microsoft and Google accounts using different authentication attacks. Kali365 conducts Microsoft device code phishing attacks, while VoidProxy acts as an adversary-in-the-middle (AiTM) reverse proxy to intercept Google sign-ins and steal authentication session cookies.
Truffle Security has launched the Secrets Revocation Program, a free initiative that helps API key issuers revoke 1.4 million publicly exposed secrets that remain valid. Since 2022, the company has discovered secrets embedded in public repositories, packages, container images, datasets, and applications. Earlier this year, Truffle Security reported that it had found 543,000 valid credentials leaked in public GitHub repositories. In a separate investigation, the company identified more than 431,000 AWS secrets across code repositories, Git histories, datasets, Docker images, registries, and CI logs. These secrets are typically exposed by mistake during the development process, by committing credentials to a public repository, including them in published Docker images or packages, or leaving them in configuration files uploaded with an app or dataset.
DigiCert warns it suffered a data breach after a server migration mistakenly exposed an internal business system to the internet, allowing unauthorized access to customer information and identity documents. The certificate authority says the breach affected a small portion of its archived customer data, including names, business email addresses, phone numbers, and business addresses. For some people, the exposed records also contained personal identification documents, such as driver's licenses, passports, and utility bills. DigiCert has begun sending data breach notifications to affected customers, with cloud architect Fabian Bader sharing a copy of the notification on X. "Thanks @digicert that your data leak was 'limited' to only my full PII data and potentially any verification documents I submitted for my certificate," Bader wrote.
Attackers target IT service desks to bypass MFA by impersonating employees and convincing staff to reset passwords or register new authentication devices. Specops Software explains why traditional identity verification methods are no longer sufficient and how organizations can secure account recovery against social engineering risks.
Cybersecurity researchers have disclosed details of a now-patched vulnerability that could be exploited to bypass Lockdown Mode protections in OpenAI ChatGPT. Lockdown Mode was introduced by the artificial intelligence (AI) company in June 2026 as an optional setting to minimize data exfiltration risks arising from prompt injections. "The flaw could enable silent chat data theft, phishing,
October 2026 brought another wave of cyber threats targeting US businesses, with attackers exploiting trusted platforms, legitimate authentication processes, and everyday business workflows. For enterprises, these campaigns raise the stakes beyond individual security incidents, increasing the risk of financial losses, data breaches, and costly operational disruption. Below, we examine five
Apple has revealed that it plans to start checking public iCloud links for harmful content starting with iOS 27.3, iPadOS 27.3, macOS 27.3, and visionOS 27.3. The change specifically applies to scenarios where users share iCloud content, such as photos in iCloud Photos or files in iCloud Drive, by sharing a link and making it accessible to "Anyone with the link." When users choose this option,
Zyxel GS1900 network switches that received the vendor's earlier fix for an actively exploited flaw are still open to the exploit attackers used, researchers at Binarly found. The exploit lets anyone who can reach a switch's web management interface run commands on it without logging in. Zyxel has released new firmware for 10 GS1900 models that closes the hole. The earlier fix is the 2.90
Confirm this action.
Leaving now will discard your changes.