NSO Group confirmed to TechCrunch that an unnamed group of American investors has taken “controlling ownership” of the surveillance tech maker.

The fragmentation of EU programmes needs to be reduced to help companies scale up.

Tilde has launched a powerful open-source language model trained on Europe's LUMI supercomputer.

When developing AI systems, practitioners often focus on model building, while sometimes underestimating the importance of analysing the diverse data collection mechanisms. However, the diversity of m...

Stakeholders are encouraged to provide feedback and examples on the interplay with other reporting regimes in this public consultation.

ENISA is the EU agency dedicated to enhancing cybersecurity in Europe. They offer guidance, tools, and resources to safeguard citizens and businesses from cyber threats.

To justify the obscene amounts of money they have raised, AI firms are trying to convince people that their chatbots can do a lot more for you than just write emails and generate studio Ghibli images ...

Memory Integrity Enforcement (MIE) is the culmination of an unprecedented design and engineering effort spanning half a decade that combines the unique strengths of Apple silicon hardware with our adv...

An attacker installed Huntress onto their operating machine, giving us a detailed look at how they’re using AI to build workflows, searching for tools like Evilginx, and researching targets like softw...

The Federal Trade Commission is issuing orders to seven companies that provide consumer-facing AI-powered chatbots seeking information on how these firms measure, test, and monitor potentially nega

AI agents can be very powerful tools, but they also can increase the risks your organization faces if not properly secured. Here’s what you need to know.

NIST has released a concept paper and proposed action plan for developing a series of NIST SP 800-53 Control Overlays for Securing AI Systems, as well as a launching a Slack channel for this community...

The European Commission has signed an agreement today with ENISA, the European Union Agency for Cybersecurity for the operation and administration of the EU Cybersecurity Reserve.

The discovery of PromptLock shows how malicious use of AI models could supercharge ransomware and other threats.

The startling lack of good security practices around AI has cybersecurity veterans wondering which decade we're living in.

Discover how prompt injections can lead to zero-click exploits threatening AI agents built using Copilot Studio. Learn about real-world risks, including data leakage and security blind spots. Bypass C...

An internal Meta policy document reveals the social-media giant’s rules for chatbots, which have permitted provocative behavior on topics including sex and race.

The Cleveland Clinic and startup Piramidal are developing an AI model trained on brain wave data to monitor intensive care patients.

The goal of this wristband is to provide less invasive tools to interact with computers for people with motor disabilities.

Four related algorithms are now ready for use to protect data created and transmitted by the Internet of Things and other electronics

Hallucinations, Hidden Costs & Human-in-the-Loop Realities

Carnegie Mellon researchers show how LLMs can be taught to autonomously plan and execute real-world cyberattacks against enterprise-grade network environments—and why this matters for future defenses.

Perplexity is repeatedly modifying their user agent and changing IPs and ASNs to hide their crawling activity, in direct conflict with explicit no-crawl preferences expressed by websites.

Es el momento de repolitizar la cultura científica como un acto de resistencia democrática y exigir a la comunidad científica una dosis de activismo

The Template annexed to this Explanatory Notice aims to provide a common minimal baseline for the information to be made publicly available in the Summary of Training Content for general-purpose AI mo...

Ameneh Dehshiri explains how the EU AI Act can increase transparency in GPAI training data and potentially strengthen GDPR enforcement.

New research from the University of Waterloo’s Cybersecurity and Privacy Institute demonstrates that any artificial intelligence (AI) image watermark can be removed, without the attacker needing to kn...

Vendetect is our new open-source tool for detecting copied and vendored code between repositories. It uses semantic fingerprinting to identify similar code even when variable names change or comments ...

A scalable file analysis and data generation platform that allows users to easily orchestrate arbitrary docker/vm/shell tools at scale. - cisagov/thorium

CPUs provide isolation mechanisms like virtualization and privilege levels to protect software. Yet these focus on architectural isolation while typically overlooking microarchitectural side channels,...

Our new report finds that teens' AI companion use is commonplace, complicated, and sometimes concerning.

Confronting the production-progress paradox

2025 Copyright Innovation Challenge

The OpenID Foundation membership has approved OpenID for Verifiable Presentations 1.0 as an OpenID Final Specification. ·

By Paddy Leerssen, Anna van Duin, Iris Toepoel, and Joris van Hoboken Discussion of DSA enforcement tend to focus on regulatory action by the European Commission and national Digital Services Coordina...

The Commission has adopted a delegated act outlining rules granting access to data for qualified researchers under the Digital Services Act (DSA).

Methodology to perform an AI System Assessment

AI is reshaping code security—shifting metrics, unknown bugs, and autonomous decisions humans may never understand.

If the United States wishes to compete in cyberspace, it must compete against China to secure its offensive cyber supply chain.

Critical vulnerabilities in OneLogin's AD Connector leaked authentication credentials, enabling account impersonation.

Global leader in hands-on learning for enterprise and cloud security education. Join 10000+ infosec professionals from 130+ countries