Attention: You are using an outdated browser, device or you do not have the latest version of JavaScript downloaded and so this website may not work as expected. Please download the latest software or switch device to avoid further issues.
| 19 Jul 2026 | |
| Written by Gabi Gerber | |
| Attacks & Threats |
| Hacking Topics, Security Operation Center |
Ernst & Young has begun notifying clients of a data breach after an attacker compromised a third-party support platform and downloaded documents containing personal and financial information used in preparing tax returns.
The incident affected an external IT service-management system used by EY employees supporting the firm’s client tax services. Support tickets created through the platform could include document attachments, creating a route through which confidential tax records and related financial information were exposed.
EY detected suspicious activity involving the platform on April 23, 2026, and launched an investigation with assistance from an independent cybersecurity firm. The investigation determined that an unauthorised third party had accessed the system between March 28 and April 12 and downloaded documents belonging to multiple clients. More here
Dark Reading Confidential Episode 20: Expert Rich Mogull reflects on lessons cyber teams should pull from the OpenAI agent's attack on Hugging Face. More...
New research shows how attackers can use security alerts and blocked events to manipulate and hijack AI agents. More...
A study of more than 6,000 patches found that even working patches can introduce new bugs, break something else, or are … More...
Advanced LLMs escaped their sandboxes while attempting to achieve a non-malicious benchmark test objective. More...
A myriad of software makes up the typical AI harness, and trust issues between the components can create concerning atta… More...
Dark Reading Confidential Episode 20: Expert Rich Mogull reflects on lessons cyber teams should pull from the OpenAI agent's attack on Hugging Face. More...
New research shows how attackers can use security alerts and blocked events to manipulate and hijack AI agents. More...
A study of more than 6,000 patches found that even working patches can introduce new bugs, break something else, or are … More...
Advanced LLMs escaped their sandboxes while attempting to achieve a non-malicious benchmark test objective. More...
A myriad of software makes up the typical AI harness, and trust issues between the components can create concerning atta… More...