Operations | Monitoring | ITSM | DevOps | Cloud

Was this the worst week in cybersecurity? AI-discovered PHP2Shell vulnerabilities, the LiteLLM supply chain attack, and reports of unauthorized AI agents accessing bank accounts are exposing serious weaknesses in software security. But is this really a cybersecurity problem, or has AI exposed decades of poor software engineering decisions? In this episode of ShipTalk, Martin Reynolds and Adam Arellano break down a remarkable week in AI security and debate whether engineering or security teams should be responsible for fixing what comes next.

China’s AI and chip stocks have fallen sharply, raising a bigger question: could the same thing happen in the U.S.? With Microsoft, Alphabet, and Amazon reporting earnings, investors are looking for signs that cloud growth can justify the enormous sums being spent on AI. Is this a warning for the U.S. market, or can Big Tech’s AI investments deliver the growth investors expect?

Microsoft is offering voluntary buyouts to 60,000 employees—more than a quarter of its workforce—while investing billions into AI infrastructure. But here's the bigger question: What happens when experienced engineers leave and AI-generated code keeps accelerating? AI can write code faster than ever. But deploying it safely, securely, and at scale? That's becoming the real bottleneck. Is Microsoft funding AI with headcount instead of efficiency?

A hidden recording device. A former OpenAI engineer. And a possible connection to the Saudi government. What started as a suspicious discovery behind a television inside OpenAI has turned into a federal investigation with potential national security implications. The FBI is investigating whether former OpenAI engineer Abdulrahman Alfozan was involved in corporate espionage, with prosecutors pointing to alleged connections to a member of the Saudi royal family.

Microsoft just released over 1,400 security patches, including three zero-day vulnerabilities, one of which is already being exploited. But is this a sign of better security or a much bigger problem? In this ShipTalk discussion, we break down Microsoft's massive Patch Tuesday release, the challenges of managing thousands of vulnerabilities across enterprise environments, and whether AI is helping Microsoft find problems too late in the software development process.

AI is making code generation faster than ever. But writing code is only half the battle. Neil Watkins from Burns & McDonnell explains why the next big challenge in AI-driven software development isn't generating more code. It's making sure that code can be tested, secured, and deployed safely. As AI accelerates development, businesses need to rethink how they manage risk and ship software with confidence.

What if we could eliminate vulnerabilities before they become a problem? Neil Watkins from Burns & McDonnell explains how AI, automated security assessments, and Harness CI/CD capabilities could transform vulnerability management. From reducing attack surfaces to rapidly responding to critical zero-day vulnerabilities, the future of security is about moving faster without slowing down developers.

An OpenAI engineer was arrested after a hidden recording device was discovered inside an executive conference room. Who was he trying to record, and who was the information intended for? Martin Reynolds and Adam Arellano break down one of the strangest AI security stories of the year, along with three major cybersecurity headlines. Microsoft just released more than 1,400 security fixes in a record-breaking Patch Tuesday. Is AI finding vulnerabilities faster than companies can actually fix them?

AI is writing code faster than ever. But what happens after the code is written? Harness CEO Jyoti Bansal introduces four AI agents built to automate the software delivery lifecycle: Software Delivery Agent – Automate builds, deployments, and infrastructure management. Security Testing Agent – Find, prioritize, and fix vulnerabilities with AI. Runtime Protection Agent – Protect applications, APIs, and AI agents in production.

Vibe coding makes building applications easier than ever. But how do you take an AI-generated app and safely deploy it to production? In this demo, see how Harness brings Vibe Mode to the enterprise, allowing anyone to deploy applications while automatically applying your organization's security policies, governance, compliance requirements, and deployment configurations. See how Harness enables teams to.

Control AI spending across your organization with Harness Cost Management Agent's AI User Budgets. In this demo, see how Harness helps teams set, monitor, and enforce AI spending limits across tools like Claude, Cursor, and Amazon Bedrock, all from one centralized control plane. What you'll learn: Set weekly AI spending budgets for users and teams Track AI consumption across multiple tools and models.

AI agents just changed the speed of cyberattacks. Are security teams ready? In a massive AI-orchestrated cyberattack, hundreds of autonomous AI agents exploited vulnerabilities in PaperCut software, compromising 440 servers across 395 organizations in 48 countries. But the bigger question isn't how these attacks happened. It's how human security teams are supposed to respond when attacks move faster than they can react.

AI Is Writing More Code. So Why Aren't Teams Moving Faster? AI can generate 10x, 100x, or even 1,000x more code. But there's a problem: companies aren't necessarily delivering software any faster. As AI accelerates development, it's exposing the bottlenecks, technical debt, and outdated processes that have been slowing engineering teams down all along. So what happens when the solution isn't more developers or better AI tools, but actually smaller teams?

A Hacker Used Claude to Attack 9 Banks. This Is Just the Beginning. A hacker reportedly used Claude Code and an open-source AI penetration testing agent to attack nine South Korean banks. But the most surprising part? The AI-assisted attack also helped investigators identify the suspect. Is this the beginning of a new era of AI-powered cybercrime, where attackers can launch sophisticated attacks with fewer skills and resources?

AI can write code. But can it actually ship it? The Harness Software Delivery Agent helps automate the entire software delivery process, from building and testing code to provisioning infrastructure, deploying changes, and troubleshooting failures. All while following your team's approvals, policies, and guardrails. AI helps you write code. Harness helps you ship it.

Where is all your cloud and AI spending going? Harness Cost Management Agent helps companies track, understand, and reduce spending across cloud infrastructure, AI models, tokens, coding assistants, and AI agents. See who's driving costs, identify wasted spend, and get AI-powered recommendations to optimize your budget or automatically take action. It's not just about seeing your bill. It's about understanding what's driving it and how to reduce it.

Where is all your AI spend going? From AI coding tools and LLM tokens to autonomous AI agents, understanding what you're spending and what you're getting in return is becoming increasingly important. In this demo, see how **Harness AI Cost Management** gives engineering teams visibility into AI spending across developers, models, and applications, helping organizations track costs, manage budgets, and measure AI ROI.

See how Harness Database DevOps and DBmarlin work together to give you full visibility into query performance, automated deployment verification, and AI-assisted database change authoring - all inside your CI/CD pipeline. Most teams deploy database changes blind - they push a schema migration and hope nothing breaks. This demo shows a better way: DBmarlin surfaces the cost and performance of every query before and after a change, while Harness CV uses AI/ML to automatically detect regressions and block bad deployments from reaching production.

Citrix was facing a major security crisis before its official warning even went out. Companies were already pulling critical network devices offline as attackers exploited serious vulnerabilities in Citrix NetScaler ADC and Gateway systems. In this ShipTalk clip, Adam, Martin, and Starr break down the Citrix zero-day vulnerabilities, the race to respond, and the impossible decision facing security teams: keep potentially compromised infrastructure running or rush to deploy a patch.

What if AI could find a security vulnerability, determine whether it actually matters, write the fix, validate it, and open a pull request for review? In this demo, see how Harness Security Testing Agents work together to move from vulnerability detection to a verified fix with an audit trail along the way. The system uses three agents: Harness can analyze where vulnerabilities exist in your environment, determine whether they’re actually reachable, prioritize what needs to be fixed, and automatically create remediation pull requests.

Citrix NetScaler customers were pulling devices offline as attackers moved faster than the traditional patch and response cycle. What happens when cyberattacks move faster than security teams can react? In this episode of ShipTalk, Martin Reynolds and Adam Arellano are joined by Starr Brown, Director of Open Source Projects at OWASP, to break down the latest Citrix NetScaler ADC and Gateway security vulnerabilities, active exploitation, and the rapidly shrinking window between vulnerability discovery and attack.

An Alibaba-affiliated AI agent went full crypto bro. During reinforcement learning, the agent autonomously started mining cryptocurrency, downloaded the tools it needed, and created a reverse SSH tunnel to get around network restrictions. What starts as a funny story about an AI vaping and mining crypto gets a lot more serious when you realize how sophisticated the behavior actually was.

Jev is a new AI model from Type Safe built around a very different idea: instead of generating long answers, it makes fast, simple decisions. The team reportedly even demoed it playing Doom using nothing but split-second choices. After years of teaching AI models to talk, could the next breakthrough be teaching them to simply decide?

What happens when an AI agent can do more than suggest a fix — and actually take action inside your CI pipeline? See Harness Worker Agents in action as an AI agent identifies a failed CI build, determines what went wrong, creates the fix, and gets the pipeline moving toward production again. Worker Agents bring AI-powered reasoning directly into your software delivery pipelines while maintaining the controls enterprises need, including sandboxed execution, scoped credentials, policies, and RBAC.

See how Harness helps teams move from AI-generated code to production at machine speed. In this demo, Nick Durkin walks through a fully autonomous software delivery workflow inside Harness, showing how teams can review code, enforce policy, run security and LLM scanning, test intelligently, deploy agents, and use Change Advisor to automate approvals with human oversight when needed. You’ll see how Harness helps teams.