Browse All

Over 600 Google employees, including many from DeepMind (Google's AI research lab), signed a letter asking CEO Sundar Pichai to prevent the Pentagon from using Google's AI models for classified purposes (secret military projects). The employees argue that the only way to ensure Google isn't associated with potential harms from such uses is to reject these classified projects entirely, since otherwise they could happen without employee knowledge or oversight.

This paper describes a security problem in blockchain payment channels (like the Lightning Network, which allows faster transactions by bundling multiple payments together): malicious intermediate nodes can intercept funds by reading payment conditions sent in plaintext. The authors propose a solution using a new encryption method called CUAP-PRE (ciphertext unlinkable autonomous path proxy re-encryption, which encrypts payment instructions so intermediate nodes can't see or trace them) combined with an improved payment protocol that lets the final receiver control decryption rights in reverse order to unlock the funds.

LVMark is a watermarking technique that hides invisible digital markers in video diffusion models (AI systems that generate realistic videos from text descriptions) to protect against unauthorized use. The method improves on existing approaches by maintaining temporal consistency (keeping the watermark stable across consecutive video frames) and preserving video quality through a specialized decoder that combines frequency analysis with color information. The technique successfully embeds 512-bit capacity watermarks that remain detectable even when videos are distorted or attacked.

ShadowCoT is a backdoor attack (a hidden vulnerability inserted into an AI model that causes it to misbehave when triggered) that targets Chain-of-Thought reasoning, which is a technique where LLMs show their step-by-step thinking to solve complex problems. Unlike simpler attacks, ShadowCoT hijacks the model's internal reasoning process by subtly rewiring how attention flows through the model and changing intermediate representations (internal data the model creates while processing), allowing it to produce logical-sounding but harmful outputs while avoiding detection.

This paper presents a new method for securely transmitting data between two users over a wireless channel where an eavesdropper might be listening. The researchers use polar coding (a technique for encoding data reliably) combined with feedback (information sent back from receiver to sender) to create correlation between transmitted signals and add dummy messages that confuse eavesdroppers. The proposed scheme achieves both reliable communication and strong secrecy without requiring simplified assumptions about the channel.

Sanitizer is a new framework for federated learning (a training method where multiple computers learn together without sharing raw data) that addresses two major challenges: privacy and Byzantine attacks (when some computers send bad or malicious data). The framework uses more efficient cryptographic techniques and two new protocols for combining data from multiple sources, achieving 19-23 times faster performance than existing methods while maintaining accuracy and security.

This research presents HEFLGuard, a security system that detects backdoor attacks (hidden malicious behavior injected into AI models) in homomorphic encryption-based federated learning (a privacy-protecting technique where multiple parties train an AI model together using encrypted data). HEFLGuard works by having the server create validation models from encrypted client groups and having clients compare their local model outputs on trusted samples to identify suspicious behavior, while also using Byzantine fault-tolerant aggregation (a method that functions correctly even when some participants act maliciously) to ensure robustness.

Deepfake voice and video attacks (AI-generated replicas of real people) are becoming increasingly common and costly, with tools that require only three seconds of audio and cost almost nothing to create. Attackers target finance employees and IT staff by impersonating executives on calls or video meetings to authorize large money transfers or credential changes, and these attacks bypass traditional security tools because they rely on tricking people rather than exploiting software vulnerabilities. Organizations that have successfully stopped these attacks all used the same defense: training employees to pause and verify requests before acting on them.

Some people worry that advanced frontier LLMs (large language models, AI systems trained on massive amounts of text) like Claude Mythos and GPT-5.5 could cause serious cybersecurity problems by being misused for attacks. However, security researcher Ari Herbert-Voss suggests this situation could also present opportunities.