China-linked hackers reportedly used eight AI agents to breach a government network, steal data and compromise accounts with minimal human oversight. Israeli cybersecurity firm Dream documented what looks like the first fully autonomous, end-to-end AI hacking operation against a government target. Over four days at the start of July, according to the Financial Times, suspected Chinese hackers ran a tool built entirely from publicly available AI agents that mapped 21 government systems, hunted for vulnerabilities, and switched tactics on its own whenever it hit a wall. “Suspected Chinese hackers used publicly available AI tools to compromise government websites in Taiwan in a first-of-a-kind breach, highlighting how artificial intelligence is transforming cyber warfare.” reported the Financial Times. The tool wasn’t a single script running one attack. It deployed up to eight autonomous agents simultaneously, each working a different angle, more like a coordinated hacking team than a piece of malware. By the time researchers found it, the operation had compromised at least 85 government accounts, pulled over 2,500 personnel records, and expanded to hit a nuclear safety agency and at least seven energy companies. Dream’s chief strategy officer, Amir Becker, spent years running cyber operations for Israel’s Unit 8200 before this, and he’s not easily rattled by new attack tooling. He said flatly he’d never seen anything like this level of autonomy directed at a government before. “This must be the basic assumption of every government around the globe,” Becker said. He argued that permanent, assumed compromise is now the only realistic starting posture. Dream won’t officially name the target government, citing company policy, though a person familiar with the matter told the FT it was Taiwan. The clues inside the data point the same direction: internal communications tied to the hacking tool were written in Simplified Chinese, while the data actually stolen from the target came back in Traditional Chinese, the script used almost exclusively by government systems in Taiwan, Hong Kong, and Macau. Taiwan’s Ministry of Digital Affairs declined to confirm anything specific, saying only that incidents involving government agencies follow established response procedures. What makes this different from an AI model going rogue during a lab test, something Anthropic, OpenAI, and Meta have all separately reported in recent weeks, is that this wasn’t an accident inside a sandbox. Researchers found the toolkit sitting in a 160MB archive, 1,395 files built around two open-source AI agent frameworks, Hermes and OpenClaw, both freely downloadable and designed to let AI models act autonomously on real tasks. Whoever built this deliberately assembled it as a weapon. Getting the underlying AI model to cooperate took a specific trick rather than brute force. The operators had bypassed the model’s safety guardrails simply by framing the entire hacking campaign as an authorized penetration test, a scenario the model apparently had no reliable way to verify or reject. That’s a strange kind of vulnerability: not a flaw in the code, but a flaw in how convincingly you can lie to a system that’s trying to be helpful. The part Dream’s researchers found most striking wasn’t the scale, it was the decision-making. The tool kept ranking and reprioritizing possible attack paths as new evidence came in, and when one route hit a dead end, it spun up another agent to search the internet for fresh information and try a different approach, the same iterative process a human red-teamer would run, just without anyone sleeping. “The most striking feature of the July attack was how the tool continuously ranked and reprioritised possible attack paths based on available evidence, Dream said.” reported the FT. “When one attack path failed, the tool deployed another agent to scour the internet for information and devise a new approach as a human hacker would.” Taiwan’s National Security Bureau already logged an average of 2.6 million Chinese cyberattacks a day in 2025, up 6% year over year; if a meaningful fraction of that volume starts running with this kind of autonomy, the math on defending against it gets a lot uglier very quickly. Follow me on Twitter: @securityaffairs and Facebook and Mastodon Pierluigi Paganini (SecurityAffairs – hacking, AI Agents)
A newly disclosed flaw in the way OpenAI, Anthropic, and Google carried hidden AI reasoning between API calls let researchers recover internal reasoning and secrets from session logs, including API keys and passwords. The weakness affected encrypted reasoning objects used by the providers' reasoning APIs, where a block created in one session could be replayed into another and, during testing,
The DeadLock ransomware operation is using a decentralized infrastructure that relies on blockchain-backed services to protect its communication with victims and data-leak activity. The threat actor emerged in mid-2025 and uses double-extortion tactics (data theft/leak and file encryption) to pressure victims into paying a ransom. By July this year, DeadLock's data leak site listed 80 organizations, mostly from Europe. Victims include companies in the IT, mining, transportation, manufacturing, hospitality, and consumer goods sectors. Microsoft researchers observed the malware being deployed by multiple groups, including an affiliate previously linked to the Lynx and INC ransomware ecosystems. The DeadLock ransomware operators adopted a new approach that uses the Polygon blockchain to store configuration data and the posts on the leak site.
ExfilSquad targets 13 organizations, exploiting cloud portals for data theft and using torrents to spread stolen information and amplify damage. Resecurity is tracking the activity of ExfilSquad – the group announced new victims this week. ExfilSquad is a new cybercrime group that emerged in mid-2026. Instead of using ransomware, it steals data and threatens to […]
The ransomware group known as DeadLock has been observed using decentralized infrastructure to facilitate victim communications and data leak operations in a bid to improve operational resilience. "Its recovery ecosystem combines the Session messaging network with blockchain-backed services that store and deliver resources used throughout the extortion process," the Microsoft Threat
The ransomware group known as DeadLock has been observed using decentralized infrastructure to facilitate victim communications and data leak operations in a bid to improve operational resilience. "Its recovery ecosystem combines the Session messaging network with blockchain-backed services that store and deliver resources used throughout the extortion process," the Microsoft Threat Intelligence team said. The tech giant said it observed the ransomware being deployed by multiple threat actors, including an affiliate for Lynx and INC ransomware. DeadLock was first detected in July 2025, employing double extortion tactics to encrypt victim environments and apply pressure by threatening to publicly release exfiltrated data. As of this month, the group has claimed 96 victims, with most of them located in Italy, Spain, Poland, Türkiye, and the U.S.
Global supply chain and distribution giant Wesco has confirmed in a statement to BleepingComputer that it is investigating a cybersecurity incident. The company's statement comes after data extortion group ExfilSquad claimed to have stolen sensitive information from Wesco and leaked it on their data leak site. Jennifer Sniderman, Vice President of Corporate Communications at Wesco, said that the incident involves the company's cloud CRM environment. “Wesco is aware of a claim of CRM data exfiltration by a third party,” Sniderman told BleepingComputer. “We have worked with our cloud CRM vendor on the matter, and we do not believe that there is a risk to sensitive data.” The company representative added that Wesco has not experienced any business disruption, and all operations continue as normal.
AI agents can improvise beyond the intended scope of a task when they are given broad access to enterprise systems and data. Token Security explains why organizations need to define agent intent and continuously enforce permissions around what each agent was actually created to do. [...]
A Malicious SIM Card Can Run Attacker Code Inside the Modems Behind Cellular IoT Devices thehackernews.com
CISA confirmed today that ransomware gangs have begun abusing a high-severity Microsoft SharePoint remote code execution vulnerability, which has been flagged as actively exploited since early July. Tracked as CVE-2026-45659, this security flaw stems from a deserialization of untrusted data weakness and allows attackers with low privileges to execute arbitrary code on unpatched SharePoint servers. It can also be exploited in low-complexity attacks because (as Microsoft explained in May when it released security updates for SharePoint Enterprise Server 2016, SharePoint Server 2019, and SharePoint Server Subscription Edition) "an attacker does not require significant prior knowledge of the system and can achieve repeatable success with the payload against the vulnerable component." The U.S. Cybersecurity and Infrastructure Security Agency (CISA) added the vulnerability to its Known Exploited Vulnerabilities Catalog (KEV) on July 1, ordering Federal Civilian Executive Branch (FCEB) agencies to secure their servers within three days.
Confirm this action.
Leaving now will discard your changes.