New Research Exposes Weaknesses in Tokenizers Used by Major LLMs Rashmi Ramesh (rashmiramesh_) • May 7, 2025 Researchers have revealed that adversaries can bypass tokenizers using various subtle methods, including emojis and zero-width spaces. (Image: Shutterstock) Recent findings indicate that sophisticated obfuscation tactics can effectively circumvent the safety mechanisms employed…
Cybercrime, Events, Fraud Management & Cybercrime FBI’s Sanjay Virmani on Recent Takedowns Tom Field (SecurityEditor) • May 2, 2025 Sanjay Virmani, Special Agent in Charge, FBI, San Francisco Establishing strong collaborations with private sector entities and international partners is crucial in the fight against cybercrime, emphasizes Sanjay Virmani, the Special…
Artificial Intelligence & Machine Learning, Events, Next-Generation Technologies & Secure Development Ken Huang of DistributedApps.ai Discusses Agentic AI Risks and Threat Modeling Rahul Neel Mani (@rneelmani) • May 2, 2025 Ken Huang, Chief AI Officer, DistributedApps.ai In an era where AI agents are increasingly autonomous and equipped with responsive tools,…
San Francisco, United States, April 29th, 2025, CyberNewsWire By integrating agentic AI with contextual threat intelligence, SecAI revolutionizes the threat investigation process, transforming it from a bottleneck into a powerful force multiplier. SecAI, an AI-driven threat intelligence firm, made its official debut today at the RSA Conference 2025 in San…
Verizon Business Unveils Insights from the 2025 Data Breach Investigations Report Verizon Business has released its highly anticipated 2025 Data Breach Investigations Report, which evaluates over 22,000 security incidents, including 12,195 confirmed data breaches. This report highlights that credential abuse and vulnerability exploitation remain the predominant attack vectors, accounting for…
OpenText Unveils Plans for Enhanced Threat Detection and AI Integration In a recent announcement, OpenText, a leader in information management solutions, detailed its strategic commitment to enhancing cyber threat detection and response capabilities through its executive vice president Muhi Majzoub. This initiative will focus on incorporating security features directly into…
Surge in Data Uploads to Generative AI Platforms Raises Concerns Over Security Risks Recent findings from the cybersecurity firm Netskope indicate a dramatic surge in the amount of sensitive internal business data uploaded to generative AI applications, revealing a startling 30-fold increase over the past year. This influx includes a…
Recent findings by cybersecurity researcher Fowler reveal alarming content associated with the GenNomis website, which had previously hosted a database containing disturbing AI-generated imagery. Among the material identified were not only questionable adult images but also potential child sexual abuse material (CSAM) linked to face-swapping technologies. Fowler indicated that several…
Research conducted by F5 reveals that the proliferation of generative AI has significantly transformed the dynamics of bot traffic, indicating that automated users now surpass human users in accessing informational web content. The findings are part of the 2025 Advanced Persistent Bots Report, which analyzed over 207 billion web and…
