Codex Security Agent, AI warfare, US Cyberfare, Threat actors use AI, CyberStrikeAI

If you enjoy our newsletter, please consider to be a paid subscriber to help us keep more news and updates coming out.

Notice: The Book Report Q4, 2025 is available! Download here.

Highlights

• Codex Security Agent

• AI warfare

• US Cyberfare

• Threat actors use AI

• CyberStrikeAI

---

Deep Dive

Codex Security Agent OpenAI

• Finds complex vulnerabilities

• Provides actionable security fixes

• Uses AI agentic reasoning

• Builds project threat models

• Validates issues in sandbox

  EY 4T SQL online, Aardvark, Atlas’s Omnibox, OpenAI releases “safeguard”, false reports of Gmail breach

AI warfare The Guardian

• Iran conflict shows paradigm shift

• AI speeds battlefield decisions

• Identifies military targets

• Recommends weapons choices

• Evaluates strike legality

• Human oversight weakening

Cyber Model Arena, AI adversarial use, Project Zero, Big Sleep, Cyber Pulse Report

US cyberspace WhiteHouse

• Adopt zero-trust architecture

• Deploy post-quantum cryptography

• Use AI-powered cyber defense

• Protect critical infrastructure systems

• Secure energy and telecom networks

• Build national cyber workforce

Threat actors use AI Microsoft

• Reduces technical attack barriers

• Researching CVEs with LLMs

• Reverse proxies and tunneling setups

• Deepfakes enhance impersonation attacks

• Voice cloning enables vishing scams

• Prompt injection attacks increasing

CyberStrikeAI BleepingComputer

• Used in FortiGate hacking campaign

• Open-source AI security platform

• Uses AI orchestration engine

• Supports GPT, Claude, DeepSeek

• Firewalls and VPNs targeted

• Automates cyberattack workflows

[Available] Book Report Q3, 2025

Available: Q2 2025 Report

Available: Q1 2025 Book Report

Teaser: Y2 GenAI Safety and Security

David Sachs & what's next

How David Sachs thinks about AI & Crypto

How upcoming Trump senior AI policy advisor thinks