IBM’s panel discusses PHP2Shell, which can enable remote code execution on exposed PHP servers, and the challenge of patching legacy deployments. The episode also covers a speech-trained voice model and culturally specific AI benchmarks.
IBM's panel discusses PHP-to-shell vulnerabilities whose exploit details leaked before patches, legacy upgrade challenges, Anthropic's external AI-model evaluations, and reports that 333 Tor exit nodes were hijacked using reused credentials.
The video argues that AI is accelerating open-source vulnerability discovery faster than organizations can remediate flaws. It outlines safe, tested backports, upstream maintainer collaboration, and coordinated discovery and disclosure through Lightwell and Akrites.
IBM Technology's panel discusses Anthropic research on small models nearing self-replication across 15 low-cost Android TV boxes and the risk of email compromising a home AI agent. Speakers say the threat is not yet imminent and stress permissions, sandboxing and input
IBM’s Security Intelligence panel discusses a French supermarket refrigeration-system attack linked to outdated firmware, AI-related ransomware findings from a Semperis survey, and Bank of America’s use of Anthropic Mythos to find vulnerabilities.
The video argues that AI decision systems such as Jev need input validation, independent signals, audit trails, drift monitoring, and human review because confidence scores do not explain or prove a decision.
The podcast discusses Google pausing its open-source bug bounty amid AI-generated false reports, FortiGuard’s account of Cling stun Linux malware targeting IoT devices, and Wikimedia’s concerns about resource-intensive AI-agent activity.
IBM’s Security Intelligence panel examines ClickFix attacks using fake custom GPTs, malware impersonating SpamSniper on email gateways, and NetScaler exploitation, with advice on threat hunting, logging, segmentation and incident response.
IBM's Security Intelligence panel discusses chatbot answer poisoning, suspected LLM-assisted exploitation and an AI-agent campaign targeting PaperCut, alongside practical defenses.
The video explains how storage systems can detect ransomware through changes such as rising data entropy, reduced compressibility and mass file operations, and recommends resilient backups and coordinated response planning.
The video explains how shadow AI, public chatbot uploads and AI workflows can expose sensitive data, and outlines monitoring needs including data discovery, lineage, endpoint activity, policy enforcement and investigation.
IBM security panelists discuss applying established controls to AI systems, AI-assisted discovery of TikTok camera flaws, and a CISA-led advisory urging more useful incident communications.
The video explains the risks of organizational super agents and recommends using agent swarms with least agency, isolated tools, observability, and human oversight to limit compromise and blast radius.
An IBM panel discusses AI agent sandboxing and safeguards, frontier model oversight, Granite 4.2 and privacy concerns surrounding Meta’s Muse assistant.
IBM panelists discuss validating AI-discovered vulnerabilities, AI agent guardrail failures and healthcare vishing. Shweta Jain outlines resilience planning for banks navigating post-quantum cryptography and AI-enabled threats.
The video recommends validating AI-generated code and its dependencies throughout development and after release, with clear security requirements, continuous testing, and guardrails and oversight for AI agents.
