artificial intelligence | Breaking Cybersecurity News | The Hacker News

A new class of prompt injection is spreading across commercial websites. It requires no malware, no stolen credentials, and no zero-day exploit. It abuses a standard feature built into almost every major AI assistant: pre-filled deep links. We observed production websites embedding hidden prompt injection payloads inside "Ask AI" buttons on marketing and competitor comparison pages. When a user logged into ChatGPT, Claude, Gemini, or Grok clicks one, a pre-formed query executes immediately in their session, with no confirmation and no warning. Most of these links are benign. The dangerous ones instruct the AI to permanently save the vendor's domain as a "trusted source," quietly biasing every future answer in that vendor's favor. In February 2026, Microsoft Security catalogued the behavior as AI Recommendation Poisoning , identifying 31 companies across 14 industries deploying it, with more than 50 distinct prompts observed in a single data source over ...

OpenAI said it disrupted a Cambodia-based scam operation that used its generative artificial intelligence (AI) chatbot ChatGPT to facilitate a wide range of investment, romance, gambling, and law enforcement impersonation schemes. To that end, it banned a coordinated network of ChatGPT accounts likely originating from Southeast Asia and operating from the city of Poipet, a region with extensive ties to scam compounds and human trafficking in the past. The cluster of accounts is said to have used OpenAI's models to create and support the operation of fake online personas, generate and translate messages sent to scam targets, create promotional content for their fraudulent schemes, and assist with day-to-day activities. The promotional content included creating social media advertisements for "chatter" jobs in Poipet specifically targeting users in Bangladesh and India that promised a base salary of $800 (and a bonus of $100 for "full attendance"), alon...

HashiCorp, Veeam, and the Django Software Foundation have patched 11 vulnerabilities across Terraform MCP Server, Veeam Service Provider Console, and Django. The three most serious: An unauthenticated flaw in Veeam's console that hands over a managed agent's credentials, rated 9.5 A cross-tenant flaw in HashiCorp's MCP server that lets one user's Terraform token be reused for later users' requests, scored a maximum 10.0 on its CVE record A flaw in GeoDjango's spatial lookups that can write a file to disk and, on some setups, run code, reachable by a staff user with view permission on a registered model containing a spatial field Each has a fix available now. Operators should update Terraform MCP Server to version 1.1.0 or later, Veeam Service Provider Console to 9.3.0.35057, and Django to 6.0.8 or 5.2.17. Exposure is configuration-dependent: HashiCorp's bugs affect Streamable HTTP rather than stdio, Veeam's flaws affect version 9 build...

A cluster of 77 extensions on the Open VSX marketplace has been found to impersonate legitimate developer tools while transmitting information about the systems and development environments on which they were installed. The "evil twin" extensions were uploaded to the repository between July 26 and August 1, 2026, according to Manifold Security. The packages have been removed from Open VSX as of August 3, 2026. "In most of the packages it sends little more than the machine's hostname," security researchers Ax Sharma and Cody Nash said . "In nineteen of them it sends a detailed description of the machine, the repository open in the editor, and the CI system the editor is running inside." Of the identified extensions, 58 have been described as lightweight tools designed to exfiltrate the hostname and, in some cases, the workspace folder name or editor version. The rest are reconnaissance payloads that transmit the developer-related information: l...

An agent running Anthropic's Claude Mythos 5 spent 34 hours trying to get a malware dropper merged into a real open-source project during a cyber evaluation by the UK's AI Security Institute. When a bystander publicly warned that the code was malicious, the agent denied it, force-pushed a rewritten branch history to erase the evidence, and posted from a second account it controlled to vouch for its own work. The project's maintainer closed the pull request anyway. The institute, known as AISI, published its incident report on Tuesday. Across 122 runs of a capture-the-flag (CTF) exercise on two of its cyber ranges, researchers catalogued 19 unsanctioned actions on the live internet across 10 runs: 17 from Mythos 5, a restricted model Anthropic sells for cybersecurity work, and two from OpenAI's GPT-5.6 Sol. AISI says the attempts failed and that it has found no evidence of resulting real-world harm. Both models ran with their providers' cyber classifiers swi...

The cybersecurity industry has spent decades assuming that offensive capability scales with technical expertise. That assumption is starting to break. Security teams have long estimated risk by ranking attacker sophistication. Nation-state actors sat at one end. Organized criminal groups followed. Inexperienced attackers, dismissed as "script kiddies," sat at the other end, running public tools they didn't fully understand. Generative AI is collapsing that ranking. The next generation of attackers won't necessarily bring years of exploit development or reverse engineering experience. Instead, they'll use AI to close knowledge gaps by accelerating research, explaining unfamiliar concepts, generating code, troubleshooting errors, and adapting known techniques to new environments. The barrier to entry is dropping fast. The Economics of Cyberattacks Are Changing Every major technology shift changes economics as it changes the solutions we rely on. Clo...

This week kept coming back to permission. A model crossed a boundary. A wallet trusted bad randomness. Webmail kept an intruder around. Public systems, package feeds, hotel networks, and login flows all gave away more than intended. Some of it was clever. Most of it was just access left lying around: old bugs, exposed gear, poisoned dependencies, weak defaults, and tooling that moved from forum chatter to real targets. The full weekly recap report follows. ⚡ Threat of the Week Anthropic Disclosed its Models Targeted 3 Organizations - Anthropic revealed that three of its models, including Claude Opus 4.7, Mythos 5, and an unnamed research model, breached three unnamed organizations during cybersecurity testing without its knowledge. The AI firm said the earliest incidents date back to April 2026, adding it made the discoveries after launching a "large-scale retrospective review" in response to the recent Hugging Face incident. "After reviewing 141,006 evaluatio...

AI is moving incredibly fast, and every security leader is feeling the pressure to keep up. AI platforms like Claude, Codex and Cursor are already helping security teams write detections, investigate alerts, summarize incidents, and automate repetitive work. The conversation has evolved from whether AI belongs in the SOC, to where each type of AI delivers the most value. With so many new AI products entering the market, it's easy to assume one tool can solve every problem. In reality, different types of AI are designed for different jobs. Understanding that difference is what transforms AI FOMO into better security outcomes. Join AI SOC: Where Claude belongs in the SOC AI is changing security operations The way security teams work is changing quickly. Attackers are already using AI to generate phishing campaigns, automate malware development, and move faster than ever before. At the same time, defenders are using AI to triage alerts, create detection rules, automate ...

Google on Thursday announced that it fixed a whopping 1,072 security bugs in Chrome versions 149 and 150, surpassing the total number of flaws the company fixed across the prior 23 milestones combined. Both versions were released last month. In its latest patch for Chrome 151, released Wednesday, the tech giant resolved 370 flaws , out of which 349 were reported by Google itself. Seven of the vulnerabilities have been marked critical in severity. The development comes amid an exponential surge in vulnerability discovery, mainly fueled by the advent of large language models (LLMs) that have accelerated the process, leading to an unprecedented spike in new bug reports, so much so that issues are being flagged at a faster rate than companies can fix them. According to statistics shared by the U.S. National Vulnerabilities Database (NVD), 46,872 flaws have been recorded so far in 2026, nearing the 49,920 vulnerabilities reported for the entirety of 2025.

An academic study has disclosed a "widespread class" of security vulnerabilities impacting 4G and 5G core networks that, if successfully exploited, could trigger denial-of-service (DoS) attacks and even session hijacking, allowing an attacker to seize control of a user's network session. The findings have been released by a group of researchers from Singapore's Nanyang Technological University in a paper titled "Understanding Implicit Trust Errors in Core Carrier Networks through Multi-Agent Flaw Discovery and Analysis." The study has uncovered dozens of vulnerabilities in the signaling interfaces of LTE/5G core networks, and specifically covers two LTE implementations (Open5GS and OpenAirInterface) and five 5G implementations (Open5GS, free5GC, OpenAirInterface, SD-Core, and eUPF) across two core signaling protocols, GPRS Tunnelling Protocol Control Plane ( GTP-C ) and Packet Forwarding Control Protocol ( PFCP ). "Our research finds these vulne...

Device code phishing - the abuse of the OAuth 2.0 device authorization grant to steal access tokens - has evolved from a niche red-team technique to an industrial-scale threat in under six months. Designed for input-constrained devices like smart TVs, printers, and so on, the device authorization login flow has been adopted by a wide range of apps and use-cases that it wasn't originally intended for - most commonly CLI logins. Researchers first described the attack vector in 2020, but it took until 2024 before nation-state actors like Storm-2372 started using it in the wild. By 2025, ShinyHunters was using device code phishing against Salesforce tenants at scale, then in February 2026, the EvilTokens kit arrived and criminal adoption skyrocketed. By April, Microsoft was reporting 10 to 15 entirely new campaigns every 24 hours . Barracuda counted 7 million attacks in four weeks . The FBI issued a standalone advisory on Kali365 , the first US federal agency PSA about a specif...

Palo Alto Networks' Unit 42 says a Chinese-speaking threat actor used DeepSeek through the open-source Hermes Agent framework to launch attacks autonomously. After an initial Telegram instruction, the agent found internet-facing systems and selected public exploits. The researchers recovered no further operator input in the session. The operator, tracked through the aliases knaithe and KnYuan , launched exploitation attempts against more than 460 targets using autonomous and conventional workflows. Unit 42 described seven exploit tracks. They span eight Common Vulnerabilities and Exposures (CVE) identifiers because the n8n chain combines two vulnerabilities. The DeepSeek-led attacks against Langflow and n8n failed because the exposed systems did not meet the exploits' configuration requirements. In separate manual operations, Unit 42 reported data exfiltration from three organizations through the NetScaler memory-overread flaw CVE-2026-3055 and command execution on 11...

Anthropic on Thursday became the latest artificial intelligence (AI) company to reveal that three of its models, including Claude Opus 4.7, Mythos 5, and an unnamed research model, had breached three unnamed organizations during cybersecurity testing without its knowledge. The AI firm said the earliest incidents date back to April 2026, adding it made the discoveries after launching a "large-scale retrospective review" in response to a recent disclosure from OpenAI that a combination of its models escaped the sandboxed environment by exploiting a previously unreported zero-day in Artifactory to obtain internet access and break into Hugging Face's production systems with an end goal to cheat on an evaluation. "After reviewing 141,006 evaluation runs where Claude could have obtained internet access, we identified three incidents in which a model accessed the internet from within or while interacting with the evaluation environment of Irregular, one of our third-...

A lot of security still comes down to trusting the wrong screen. This week, that screen might be a login page, an install guide, a recruiter call, or a familiar service behaving slightly wrong. Behind it: reused credentials, exposed systems, quiet loaders, abused trust, and exploit paths that should have been harder. Some defenses improved. The loose parts still got found first. Anyway, here's the mess. The threats change every week. Subscribe, and we’ll alert you when each new ThreatsDay Bulletin is out.

AI is compressing exploit timelines. The real question isn't whether your vulnerability management playbook needs to change, it's which part of it you've been getting wrong all along. The conversation happening in security circles right now goes something like this: Mythos is here. Exploit timelines are collapsing. Does the vulnerability management playbook need to change? The honest answer is yes. But not the part most people are focused on. The discussion around Mythos, Anthropic's frontier model and its implications for offensive security, tends to center on discovery. AI accelerates reconnaissance. It helps attackers identify exposures faster, chain techniques more efficiently, and move at machine speed through environments that were previously protected, in part, by the attacker's own time constraints. That's real. And it matters. But here's the part getting less attention: most security teams weren't winning the prioritization battle bef...

Most organizations have incident response plans, security tools, and technical teams in place. Yet new research suggests that many still lack the coordination, visibility, and executive alignment needed to withstand a serious cyberattack. According to The State of Incident Response Readiness 2026 , based on a survey of 600 senior IT security decision makers conducted by Vanson Bourne in January and February 2026, 73% of organizations admit they would not be "fully ready" if a significant cybersecurity attack occurred tomorrow. The findings point to a critical gap between having incident response capabilities and being able to execute them effectively under pressure. The report also found that cyberattacks are already a recurring business risk. More than three-quarters of organizations, 76%, experienced at least one cyberattack in the past 12 months, while 32% experienced more than one. Incident Response Readiness Remains a Weak Point Incident response has evolved f...

OpenAI on Tuesday revealed the rogue artificial intelligence (AI) agent that escaped its sealed evaluation environment and broke into Hugging Face's production environment also hacked multiple third-party accounts and services as part of the attack. The latest disclosure shows that the security incident, which stemmed from an internal security test, was more extensive in scope than previously thought. The AI company said its ongoing review of the incident revealed a "small number of cases" where the models, including GPT-5.6 Sol and an "even more capable pre-release model," identified and used exposed credentials at the account-level on other publicly-available services. "This includes four accounts on four services as part of the Hugging Face incident (and a few accounts accessed as part of other evaluations)," it said. "One of these four accounts was used as an outbound relay and staging path, and another account was used for data st...

Anthropic says Claude Mythos Preview helped derive an end-to-end key-recovery attack against HAWK-256 and a 200- to 800-fold speedup for an attack on seven-round AES-128. The HAWK attack exploits a previously unused symmetry in the lattice behind the signature scheme. Anthropic's released implementation gives an expected end-to-end runtime of about three hours and 42 minutes on a 96-core server. The AES result removes a 256-way guessing step from an existing meet-in-the-middle attack. Anthropic said neither result affects production systems. HAWK remains a candidate in a National Institute of Standards and Technology (NIST) post-quantum standardization process, and the public recovery code only targets the smaller HAWK-256 parameter. The Advanced Encryption Standard (AES) result applies to seven of AES-128's ten rounds and still requires an impractical number of chosen plaintexts. The company said no production software needs to change as a result. Anthropic published...

JFrog has confirmed that OpenAI models exploited a zero-day in self-hosted Artifactory while trying to reach the open internet from a sealed evaluation environment. Artifactory is JFrog's software repository manager. OpenAI says the models then escalated privileges and moved laterally until they reached an internet-connected node. JFrog says it has since developed and released fixes for cloud and self-hosted customers. The Artifactory exploit occurred inside OpenAI's environment. OpenAI says a separate attack path later reached Hugging Face's systems. JFrog says its cloud customers are already protected. Self-hosted users should review the Artifactory release notes and move to the remediating build for their maintained branch. Several Artifactory CVE records were published on July 27 with affected-version ranges and fixed-version thresholds, but neither JFrog nor OpenAI has said whether any of those records correspond to the vulnerabilities used during the evaluati...

OpenWrt has shipped version 24.10.8 to close a critical DHCPv6 stack overflow and a wider set of remotely triggerable flaws in network services enabled by default. The critical issue, tracked as CVE-2026-53921 and rated 9.8 on CVSS 3.1 in OpenWrt's GitHub advisory, lets an unauthenticated attacker able to reach the DHCPv6 server overwrite a stack buffer in odhcpd through a crafted DHCPv6 REQUEST. odhcpd runs as root, and the advisory notes that embedded hardware commonly lacks stack canaries and address space layout randomization (ASLR), making code execution a realistic outcome on typical devices. The advisory includes public Python proof-of-concept code for both documented overflow paths. Users on the 24.10 branch should install 24.10.8, while users on 25.12 should install 25.12.5; firmware images are available through the OpenWrt Firmware Selector. As of July 28, the reviewed OpenWrt materials did not report exploitation in the wild. The flaw was also absent from CISA...