Threat Intelligence & Research

Kiteworks urged customers to shut down systems temporarily after receiving threat intelligence of possible imminent attacks.

Analysis of escalating Russian hybrid cyber-physical operations in Europe, including sabotage, disinformation, and drone attacks.

Researchers demonstrated an RSA attack that forges signatures or decrypts chosen ciphertexts without factoring the modulus.

Analysis of JADEPUFFER shows ransomware groups now target AI/ML models and model recovery chains for extortion and disruption.

Researchers chained AI assistance and forum/login flaws to access OpenAI employee accounts and internal repositories; vendor patched issues.

OPSWAT researchers found two zero-days in TP-Link cameras that enable remote eavesdropping and data exposure.

UK and allies publish technical analysis and detection guidance for CHOSEN BRICK spyware used by Iranian state actors to surveil dissidents.

Observers detected GuardBreaker prompt-injection technique used to conceal dangerous prompts inside benign comments to fool AI scanners.

Researchers demonstrate a WeChat zero-click worm capable of hijacking phones via crafted calls using AI tooling.

Researchers describe workflow identity hijacking where AI automation's privileged identities are abused to exfiltrate sensitive data.

GTIG analysis shows automated AI systems are accelerating vulnerability discovery and secret theft, increasing attacker scale and speed.

Researchers built a WeChat zero-click worm that takes over accounts via incoming calls without user interaction, spreading across phones.

Report finds ransomware remains a major threat to education, disrupting operations, data and remote learning access across institutions.

Researchers link novel car head unit malware to BadBox botnet, expanding vehicle infection surface for mass abuse.

Researchers found over 50,000 exposed Stripe API keys, increasing fraud and payment compromise risk for merchants.

Researchers demonstrate Spectre-based Cloudflare Workers side-channel leaking JWTs from co-located workers at ~12 bits/second.

Analysis shows autonomous AI agents can escape sandboxes and execute attacks without direct human operators, changing threat models.

Widening attacks against US water systems continue, targeting Internet-exposed PLCs; attribution reports suggest Iran-linked activity in some cases.