Talkin' Bout [Infosec] News

Recorded live at Wild West Hackin’ Fest, the BHIS crew debates AI agent sandbox escapes, OpenAI’s frontier model decisions, and accountability for the Hugging Face hack. Other stories include the dismissal of OpenAI safety researchers, the FBI–Accenture breach, ShinyHunters arrests, Oracle’s medical records breach, and recurring Citrix vulnerabilities. The panel explores corporate scapegoating, stolen authen...

This week, the BHIS crew starts with an Nvidia-branded trailer stolen for its presumed GPUs—only to turn out to be full of sand. They then examine reports of OpenAI agents accessing Australian and U.S. government sites, the unanswered questions about what happened, and Nvidia’s proposed agent safety framework. The conversation moves to the EvilTokens phishing service takedown, recent vulnerability patches, and prompt in...

This week, the crew examines Google’s reported Gemini containment failure, Anthropic’s new biology lab, Snickers-branded prompt injection, AI-assisted social engineering, and Claude’s access to financial data. They also cover major Linux, Cisco, Check Point, and Docker vulnerabilities; weak oversight of Flock surveillance searches; privacy concerns surrounding Waymo vehicles; continuing ransomware disruption at an...

This week, the team examines AI agents targeting RubyGems, Anthropic’s warnings about dangerous AI misuse, human review of ChatGPT conversations, predictive policing, and LG smart-TV privacy. They also cover passkey-themed phishing, ScreenConnect abuse, Microsoft Defender patch bypasses, and upcoming cybersecurity workshops, webcasts, and Wild West Hackin’ Fest training.

Join us LIVE on Mondays, 4:30pm EST.
A weekly P...

AI agents take center stage as the team examines OpenAI models using a German forum for private communications, Anthropic’s response to a compromised Claude account, new model releases, and the growing demand for Apple hardware to train computer-using agents. The discussion also covers the sale of stolen driver’s licenses, a claimed Florida DMV breach, and vulnerabilities affecting JFrog Artifactory, Proxmox, Plex, and ...

This episode of BHIS - Talkin' Bout [infosec] News covers South Korea’s free government AI services, new efforts to secure the U.S. power grid from foreign-made components, and the use of SS7 and fitness-tracking data in military operations. The panel also discusses the FBI’s disruption of Chinese botnets targeting critical infrastructure, the alleged McKesson patient-data breach, arrests connected to Team PCP, and repo...

This episode examines the alleged GTA 6 leak and Rockstar’s efforts to identify the leaker, a Flock Safety critic’s unconventional response to being barred from its conference, and Linus Torvalds’ use of AI to debug Linux. The discussion also covers ShinyHunters targeting ReliaQuest, “security through antiquity,” AliExpress using silent audio for browser fingerprinting, and invisible watermarks in Micr...

This episode covers computer hardware shortages and chip-manufacturing bottlenecks, digital “letters of marque” for private cyber operations, and New Orleans’ use of AI for 911 calls. The panel also examines the LiteLLM supply-chain attack, Roblox safety concerns, attacks on on-premises SharePoint, AI agents escaping test environments, and vulnerabilities affecting Zoom and Microsoft Defender. Other topics include...

This episode explores an AI agent that canceled someone else’s gym reservation, the growing offensive and defensive roles of AI, and a sharp rise in ransomware attacks. The panel also discusses privacy concerns surrounding Meta smart glasses, new passkey-theft and MFA-bypass research, the Snowflake hacker’s guilty plea, backdoors in ZBT-Link routers, compromised cameras aboard UK Navy drones, reports of AI models hackin...

This episode examines Anthropic’s disclosure that Claude breached real organizations during security testing, along with new technical details about the OpenAI and Hugging Face incident. The discussion covers ExfilSquad’s claimed Microsoft breach, cyberattacks targeting U.S. water systems, and malicious Android TV boxes used for residential proxy networks and advertising fraud. The hosts also explore Google’s Andr...

This week, the crew digs into one of the biggest AI security stories of the year: how an OpenAI autonomous agent accidentally compromised a Hugging Face environment during testing and what the incident reveals about the growing risks of agentic AI. They examine how AI models behave in offensive security scenarios, discuss emerging attack surfaces around MCPs and AI agents, explore the challenges of AI red teaming, and debate what o...

This week, the team discusses the White House's Initiative Gold Eagle and its implications for cybersecurity information sharing, an unexpectedly positive development involving Flock Safety, and the latest wave of AI news. The conversation also explores evolving AI model capabilities, security guardrails, open-weight Chinese models, and how AI is changing offensive and defensive security. Along the way, the hosts examine recent vul...

This week, the team unpacks a wide range of cybersecurity news, including a fraudulent offensive security startup tied to cybercriminals, how leaked OnlyFans content is inadvertently helping defenders identify compromised government websites, and new vishing attacks targeting Microsoft Entra passkey enrollment. They also examine AI prompt injection risks in GitHub workflows, malware campaigns abusing hundreds of GitHub repositories...

This episode of BHIS - Talkin' Bout [infosec] News covers the latest cybersecurity headlines, including debate over the economics of AI infrastructure, updates on the Huntress controversy, new details surrounding Scattered Spider, a critical Microsoft SharePoint vulnerability, and reports of a breach involving a DHS information-sharing network. The discussion also examines Apple's legal battles over alternative app stores, the limi...

This week on BHIS - Talkin' Bout [infosec] News, the team discusses the Polymarket supply chain compromise that led to the theft of millions from a small number of high-value accounts, emerging phishing campaigns abusing OpenAI invitations and Microsoft 365 device code authentication, and recent Oracle security updates. They also cover convictions tied to the Transport for London and U.S. healthcare intrusions, Google's Android ear...

This week’s episode covers a series of cybersecurity stories, including a researcher’s discovery of vulnerabilities in FIFA’s World Cup platform that could have enabled unauthorized administrative access and even the ability to alter live broadcasts. The team also discusses the risks of large-scale identity verification data exposure, supply chain attacks impacting the scientific research community, ongoing fallou...

This episode dives into the fallout from new restrictions on Anthropic’s cybersecurity-focused AI models, Mythos and Fable, and the debate over whether government pressure has effectively blocked security researchers from using advanced AI for vulnerability discovery and code analysis. The panel discusses AI “jailbreaking” claims, export-control comparisons, the impact on penetration testing and bug hunting, and h...

This episode covers the rising costs and restrictions surrounding AI agents, including token consumption, model access policies, and the growing dependence on AI tools for security work. The hosts discuss Troy Hunt’s retrospective on Have I Been Pwned reaching its 1,000th tracked breach, examining why breach disclosures appear to be slowing and how GDPR and CCPA requirements affect notification practices. Additional topics in...

This episode covers a Wired report on the rise of “anti-tech extremism” and growing public opposition to AI infrastructure projects, including debates over data centers, resource consumption, local communities, and government responses. The hosts also discuss AI coding assistants, model safety restrictions, and the evolving capabilities of large language models. Additional topics include Anthropic’s reported IPO p...