Experts discuss the rise of LLMjacking, where stolen AI API keys lead to massive financial losses. They explore how AI is reshaping adversary simulations, the enduring need for human expertise in the loop, and the debate over accelerating security patch timelines in the face of AI-powered threats.
Ari Weinstein discusses how Codex's 'computer use' feature allows the AI agent to operate local Mac applications in the background by combining multimodal vision with accessibility data, enabling non-intrusive, parallel task execution.
Nico Albanese from Vercel demonstrates how to build a stateful, learning AI agent from scratch using AI SDK v6. The workshop covers the core components: a tool loop, provider-executed tools like web search, end-to-end type safety, and Vercel's new persistent named sandboxes, which give the agent a file system to persist state, memory, and even self-generated tools across sessions.
95% of GenAI pilots fail due to feedback integration issues, not deployment challenges. Alessandro Cappelli argues that Reinforcement Learning (RL) provides the only systematic way to incorporate business metrics and production signals to continuously improve models, especially for complex agent-based systems.
Lloyd Blankfein, former CEO of Goldman Sachs, shares insights on leadership in crisis, the philosophy of risk management, and the evolution of institutional culture. He discusses navigating the 2008 financial crisis, the transition from a partnership to a public company, and the profound impact of technology and AI on finance and risk.
Dr. Trevor Manz from Marimo introduces Marimo Pair, an open-source agent skill that enables coding agents like Claude Code to interact with and drive reactive Python notebooks. He discusses the mechanics of agent skills, the power of recursive language models for agentic reasoning, and his work on the AnyWidget project, which bridges the gap between Python's data ecosystem and interactive web technologies.