Block CTO Dhanji R. Prasanna shares how the company is becoming AI-native. He discusses their internal open-source AI agent, Goose, which saves employees 8-10 hours weekly, how they measure productivity gains, and the organizational changes that have had an even greater impact than AI tooling.
Ben Lorica and Evangelos Simoudis discuss signals of an AI bubble, including runaway valuations and circular financing. They explore why many enterprise AI pilots fail and what separates successful adopters—namely, a focus on measurable use cases and organizational transformation—and offer a nuanced perspective on the US-China tech race, cautioning against underestimating China's progress in AI, robotics, and semiconductors.
Professor Chris Kempes of the Santa Fe Institute proposes a universal theory of life that transcends Earth-based biochemistry. He introduces a three-level framework (Materials, Constraints, Principles) to explain how abstract rules like evolution could define life on any substrate, including culture, language, or artificial intelligence, leading to the concept of convergent evolution where similar solutions, like the eye, emerge independently across different physical forms.
Co-founders Albert Gozzi and Santiago Perez De Rosso of Aleph discuss their journey building an AI-native FP&A platform. They share insights on their first-principles approach of meeting finance teams in spreadsheets, achieving rapid time-to-value, and the early challenges of building a horizontal product. The conversation covers their YC experience, scaling to a $29M Series B, and their vision for the future of finance powered by pragmatic, accuracy-focused AI.
A discussion on OpenAI's new browser ChatGPT Atlas, Andrej Karpathy's pessimistic timeline for AI agents, the DeepSeek-OCR paper on visual context compression, and a study suggesting large language models can suffer from "brain rot" when trained on low-quality social media data.
Amjad Masad, founder of Replit, joins a16z to discuss the rise of AI agents that can now plan, reason, and code for hours. He explains how reinforcement learning and verification loops unlocked long-horizon reasoning, why AI is advancing fastest in verifiable domains like code, and debates whether "good enough" AI might be a local maximum that blocks the path to AGI.