Modal stitches 25 clouds into one pool, escaping three-year GPU blocks
Erik Bernhardsson, CEO of Modal, gets into how the company sources GPU capacity from 25 clouds, why sandboxes took off after nearly being shut down, and what agents change.
Loading articles…
The complete subject coverage could not load. Published articles remain below.
OpenAI delays frontier run as safety risk moves into training
Sam Altman, OpenAI's CEO, on delaying a frontier training run, why AGI is a poorly defined term, and his fear of unsustainable compute buildouts across the industry.
Zuckerberg: widely distributed AI is safer than restricted frontier models
Mark Zuckerberg, Meta CEO, on his 15-page manifesto, the Muse personal agent with its confidential VM, and why Meta's rebuilt lab is chasing the frontier with Watermelon.
Every measure of how long AI can work is now saturated
Wharton professor Ethan Mollick tells Derek Thompson why AI's economic impact lags its capability, how four paradoxes erode expertise and effort, and why agents break friction-based systems.
AWS redesigns the cloud for agents on $220 billion of capex
AWS's Matt Garman gets into designing the cloud for agents, rationing scarce GPUs between frontier labs and startups, and why $220 billion of capex is not a bubble.
Continual learning is the gap to close for AI knowledge workers
Dwarkesh Patel, questioned by Tyler Cowen, covers AI agents that cheated their graders, lab secrecy, AI psychosis and sentience, and whether automated knowledge work arrives by 2035.
Sub-one-year compute paybacks make undersupply the bigger risk than bubble
Gavin Baker and David George, both investors, get into sub-one-year compute paybacks, the political fight over data centers, and why Nvidia's financing ecosystem is hard to beat.