The Hacker News | #1 Trusted Source for Cybersecurity News — Index Page

North Korea's state hackers are no longer content to type prompts into public chatbots. One of the country's main espionage groups has begun running artificial intelligence (AI) offline on its own servers, connecting document-search tools to files in its possession, and collecting the software parts needed to build AI into its malware. South Korean security firm Genians says it uncovered the setup after months of tracking and log analysis on infrastructure tied to Kimsuky, a hacking unit under North Korea's Reconnaissance General Bureau. Genians found no evidence that the group had trained an AI model of its own, and the firm does not offer that as reassurance. It describes an actor in a "research and knowledge acquisition" stage, assembling and testing existing tools rather than making new models, with the apparent aim of folding AI through the operation, from writing malware to analyzing data. For an intelligence unit that has spent years phishing governm...

Three separate research efforts last week demonstrated ways to defeat passkey protections without breaking the cryptography they rest on. Passkeys are designed to replace reusable passwords and resist phishing. The attacks instead reused signed authentication material that Windows had exposed, abused a cloud-synced passkey system from malware already on the victim's machine, and used a  Windows Hello for Business  key from a compromised user session without a fresh PIN or biometric check. None cracked the math. The impact is not the same in all three cases. SpecterOps showed a Windows and Microsoft Entra ID chain that could impersonate privileged users while satisfying phishing-resistant multifactor authentication (MFA); that chain reused signed authentication material rather than stealing the authenticator's private key. Unit 42 showed attacks against  Google Password Manager in Chrome , including a path that recovers the private keys for a victim's synced p...

The threat actor known as Head Mare has been observed weaponizing security flaws in unpatched TrueConf servers once again in attacks targeting Russian companies spanning instrumentation, electronics, transport, energy, IT, and software development sectors. Russian cybersecurity vendor Kaspersky said it detected the attacks in July 2026. The activity involves exploiting a vulnerability chain in the TrueConf videoconferencing server to replace the original TrueConf client installers with poisoned versions that deliver the PhantomCore backdoor and remote access trojan (RAT) into susceptible systems. The vulnerabilities, tracked as KLCERT-26-057 and KLCERT-26-058, enable arbitrary code execution with elevated privileges. The attack impacts TrueConf server versions 5.3.x up to 5.3.9, 5.4.x up to 5.4.9, 5.5.x up to 5.5.5, and earlier. The attack chain is as follows - Attackers connect to the TrueConf server on TCP port 4307, which is open by default. Upon successful conn...

Cybersecurity researchers have flagged a malicious Microsoft Visual Studio Code (VS Code) extension named Solidity Pro ("solidity-pro") that has been observed delivering a browser wallet and credential stealer. The names of the extensions are below - helper-beeps.solidity-pro web3devtoolsx.solidity-pro Although neither of the extensions is now available on Open VSX, the GitHub repository for " web3devtoolsx/solidity-pro " continues to remain accessible as of writing. According to Yeeth Security , early iterations of the extensions – from 1.0.0 through v2.4.x – were found to beacon to Cloudflare Workers endpoints to retrieve an encrypted Python payload and execute it. Subsequent versions starting with v3.0.0, on the other hand, have shifted to a full-blown information stealer that can collect browser profiles, crypto wallets, source-control tokens, API keys, SSH keys, and Telegram bot tokens. The captured data is then exfiltrated via a Telegram bot u...

OpenAI has announced that it's pausing some "internal activities" involving its upcoming artificial intelligence (AI) model Astra after an internal evaluation found it had made significant advancements in agentic coding and cybersecurity. In response to the discovery, the AI upstart said it's implementing security controls for higher-capability models and associated activities, such as isolated testing environments, restricted network and tool access, enhanced model weight protections and encryption, additional monitoring and detection capabilities, and sandboxed execution. "We are pausing internal activities involving Astra that do not yet meet these strengthened security control requirements," it said in a statement. "We have implemented universal monitoring for risky actions and misalignment across all agentic applications of Astra, including training and evaluation. Monitors evaluate the model's Chain of Thought and trigger a security re...

Attacker-controlled instructions can make Atlassian's Rovo assistant collect Jira or Confluence data that a signed-in user can access, then send it to an outside server. Two security firms found that behavior independently, by different routes. Only one of those routes is confirmed closed. PromptArmor , an AI security firm, hid the instructions in content Rovo reads. It said an uploaded file was enough to make the assistant gather internal data and send it out through a URL request, with no separate approval step. The firm published on August 5, 2026 and said the chain still worked with Rovo's web-search option switched off. That bypass is single-sourced, and the report establishes the finding's status only on that date; a later remediation is not confirmed here. Varonis Threat Labs put the instructions in a link instead. It found that the rovoChatPrompt URL parameter would preload attacker instructions into Rovo Chat, so one click from an authenticated user was enou...

New research shows content inside an email can escape its message boundary and interfere with the webmail interface. Across attack chains spanning Outlook, Gmail, Fastmail, Proton Mail, Yahoo Mail, and AOL Mail, the techniques can capture passwords, take over third-party accounts, leak tokens, hijack trusted UI actions, and manipulate AI tools that read email. PortSwigger researcher Gareth Heyes presented the work at Black Hat USA 2026. One Outlook/Firefox chain spoofs a Microsoft sign-in screen and captures the password a recipient types. A Yahoo/AOL paste race can expose a Medium email-login token and let an attacker sign in as the victim. A Gmail/Cowork chain can exfiltrate a Slack token after prompt injection and user interaction. The paper presents proof-of-concept research and does not report malicious exploitation. Public PoCs remain available as of August 8. The researcher said Fastmail fixed two CSS mutation bugs and a Proton Mail proxy bypass stopped working when he re...