Risky Business

Soap Box: Detection and response in the AI age

5 Jun 2026

The text explores the growing threat of zero-day exploits and vulnerabilities, emphasizing the need for advanced detection/response strategies, AI-driven automation in SOC tasks, collaborative AI systems for faster threat mitigation, and the evolving balance between AI capabilities and human oversight in security operations.

Risky Business #839 -- TeamPCP stole GitHub's internal repos

27 May 2026

A GitHub breach by "Team PCP" via a compromised VS Code extension exposed 3,800 internal repositories, underscoring supply chain risks, corporate underreporting, AI-driven threats, outdated dependencies, and systemic gaps in open-source and cybersecurity practices.

Risky Business #838 -- GitHub investigates possible breach

20 May 2026

Recent cybersecurity incidents, including GitHub's unauthorized access and a CISA contractor's credential exposure, highlight risks from misconfigurations, human error, legacy malware, AI-driven vulnerabilities, and enterprise tool flaws, alongside emerging threats like deepfakes, ransomware signing, and outdated infrastructure challenges exacerbated by geopolitical conflicts.

Soap Box: Where does AI fit into cloud security?

15 May 2026

Open-source cloud security tools like Prowler evolve through community contributions and AI integration, balancing automated security checks with deterministic controls amid challenges like dynamic APIs, enterprise adoption tensions, and the resurgence of foundational security measures in hybrid cloud environments.

Risky Business #837 -- GitHub Actions footgun claims TanStack

13 May 2026

Recommended: Security. Security. Security.

Summary: Cybersecurity risks from misconfigured GitHub Actions, AI-driven threats like autonomous malware, DNSSEC failures, ransomware attacks on education sectors, and challenges in AI model governance and supply chain vulnerabilities are explored, alongside discussions on regulatory responses and infrastructure resilience.

Risky Business #833 -- The Great Mythos Freakout of 2026

15 Apr 2026

Recommended: Discussion of the recent Anthropic Mythos model impact.

Anthropic Mythos AI's impact on cybersecurity, balancing its potential to accelerate vulnerability detection with debates over human expertise, polarized views on practical impact versus existential risks, and the persistence of foundational security practices amid new AI-driven challenges like patch reversal and IoT vulnerabilities.