Latest cybersecurity news covering vulnerabilities, attacks, security updates, incidents, and industry developments.
3590 articles
Imply introduced Lumi, a data platform designed to help security teams and AI agents query more security history across indexed data and object storage while retaining existing SIEM tools and workflows.
More than 700 stakeholders joined the EI-ISAC’s Election Threat 2026 forum to discuss evolving cyber, physical, and information threats and the coordination, preparation, and information sharing needed to protect election operations.
This cybersecurity roundup covers Pwn2Own exploits against Codex and LiteLLM, research on bypassing LLM safety checks, active exploitation of several vulnerabilities, a Zammad breach, and security industry developments.
A suspected ShinyHunters member was detained in Jordan while the group allegedly extorted Jeppesen ForeFlight, a former Boeing unit. Boeing is reviewing the claim; Jeppesen ForeFlight says it found no impact to operations or products.
Cloudflare says its Managed Defense beta uses deterministic data collection and specialized AI agents to correlate alerts and produce evidence-cited recommendations, while keeping analysts responsible for decisions and mitigations.
The Canto Incognito campaign used PoeLLM malware to infect more than 3,400 exposed servers, mine cryptocurrency, and scan for additional targets. The GitHub repository used to derive its C2 addresses is no longer accessible.
A 28-year-old Russian national suspected of being a core Qilin member was arrested in Osaka and extradited to Germany, where he is wanted over a 2024 ransomware attack on a logistics company.
Hadrian raised $40 million in Series B funding to expand its AI-driven offensive security platform across EMEA and the US and invest in engineering and research. Its tools map external attack surfaces and validate exploitable risks.
The FBI and Secret Service say FortiBleed remains active against internet-facing Fortinet devices, using stolen credentials to maintain access and support ransomware operations. They urge organizations to secure accounts, investigate devices, and restrict external admin
Wazuh announced a partnership with Selenium, which will use Wazuh for security monitoring, threat detection, vulnerability and compliance monitoring, and SOC services across on-premises, cloud, and hybrid environments.
ReliaQuest says IDC named its GreyMatter platform a Leader in the 2026 Worldwide Standalone AI SOC Platforms assessment. The company highlights AI-assisted SOC workflows, integrations, and governance features.
Anthropic combined its Cyber Verification Program and Project Glasswing into three access tiers for AI models, with different verification requirements and safeguards for defensive work, authorized red teaming, and safety-critical testing.
Reuters reported that 16-year-old Saif al-Din Khader, described as a ShinyHunters member, was arrested in Jordan and is in FBI custody. The article also recounts the group’s alleged attacks and a claimed breach of an FBI jobs site.
OpenSSH 10.6 introduces more frequent releases, disables LZ77 compression to reduce cross-channel attacks, fixes credential, path, and certificate handling, and enables a hybrid post-quantum signature algorithm. Earlier experimental keys must be regenerated or removed.
Tuskira’s open-source AI Agent Gateway runs in the customer’s environment, checks permissions on tool calls, and supplies stored credentials. The article also flags risks from unbound profiles, stored prompt data, and permissive demo network settings.
Anthropic is expanding its Cyber Verification Program for vetted cybersecurity teams. The company says Project Glasswing verified at least 129,000 vulnerabilities from April to July 2026, while separate research found only two of 300 reviewed flaws exploited in the wild
Wikimedia says suspected OpenAI agents tried to use its citation tool and Etherpad to fetch external data as proxies, while generating heavy traffic that may have contributed to a Wikidata Query Service outage. It found no evidence of compromise.
Sekoia says it received Spain’s ENS Alta certification, covering its information security management system and the hosting, operation, and maintenance of its SOC platform in the FRA1 region.
The FBI and Secret Service warn that the active FortiBleed campaign uses stolen credentials to compromise Fortinet firewalls and VPN gateways, lock out users, and provide access to ransomware affiliates.
Anibal Alexander Canelon Aguirre pleaded not guilty in Nebraska to charges tied to an alleged ATM jackpotting scheme using Ploutus malware. Authorities report 1,500 attacks and $40.7 million in losses; experts have not verified he developed the malware.
CrowdStrike says IDC named it a Leader in its 2026 enterprise endpoint security assessment. The company highlights Falcon Guardian’s controls for AI-agent activity and endpoint protections against malicious software supply-chain packages.
Black Lotus Labs says PoeLLM has compromised more than 3,400 servers since April by targeting open-source AI services. The malware derives changing command-and-control addresses from words in a GitHub poem.
The FBI seized two websites used to sell child sexual exploitation material, including AI-generated deepfakes, while French authorities arrested a suspected administrator. Investigators say some material came from hacked victims’ social media accounts.
Proposed federal rules would require contractors handling controlled unclassified information to meet NIST cybersecurity standards and report unauthorized access within 72 hours. The rules could take effect before the end of the current administration.
