Citizen Lab has uncovered a mercenary spyware campaign that uses targeted ads to deliver zero-click exploits to mobile devices, affecting high-profile organizations and government networks.
Anthropic researchers have demonstrated that small AI models can act as self-replicating worms, potentially creating decentralized botnets within 6 to 12 months.
Anthropic researchers demonstrated that AI models can run across networks of low-power consumer devices, raising concerns about the potential for self-replicating AI worms.
Anthropic warns that small AI models could achieve autonomous self-replication by the end of 2026, posing risks to smart devices and data center infrastructure.
NGR Softlab has integrated a new correlation rule into its Alertix SIEM to detect cyberattacks that utilize commercial LLM APIs to guide malicious activity within corporate networks.
Bill Gates warns that the immediate danger of AI is not a loss of control, but its use by malicious actors in cyberattacks, with significant negative impacts expected within the next year.
The number of malware-infected smartphones in Russia grew by 34% in summer 2026, while the effectiveness of fraudulent call blocking systems significantly declined.
MegaFon reports a 34% increase in smartphone malware infections during the summer of 2026, even as the volume of blocked suspicious calls continues to decline.
Bitdefender researchers have discovered a global distribution of smartphones with pre-installed malware that cannot be removed, affecting devices sold on major Russian marketplaces.
Hackers are using poems hosted on GitHub to manage a botnet, converting text into IP addresses to control malware and mine Monero.
Security researchers have discovered the Midnight Mimosa malware pre-installed on budget Chinese smartphones powered by MediaTek processors, preventing user removal.
Bitdefender researchers have discovered the Midnight Mimosa malware pre-installed on budget smartphones and clones, affecting thousands of devices across 150 countries.
Security researchers have analyzed a new industrial malware, S7Flip, and a related wiper tool discovered on the MalwareBazaar repository.
Security researchers have identified a new macOS malware called CloudSyncD that disguises itself as a Zoom installer to gain administrative access to user systems.
Cybercriminals are integrating AI into information-stealing malware to automatically analyze stolen data and profile victims for high-value targeting.
Cybercriminals are deploying malware with AI-driven profiling to automatically categorize victims and identify high-value targets from stolen data logs.
Researchers at Huntress have uncovered a malware campaign using malicious custom GPTs on chatgpt.com to deliver remote access trojans via an eight-stage ClickFix infection chain.
Security researchers at F6 have uncovered a new threat actor, Mimbrob, targeting Russian aerospace and IT firms with sophisticated phishing and DLL sideloading attacks.
Netskope researchers uncovered a widespread Google Ads campaign that used deceptive security warnings to target users across hundreds of websites.
The Carbonato botnet has been exploiting unauthenticated Docker APIs for nearly two years, utilizing AI agents to execute commands and exfiltrate data.
