Frontier
Frontier
Long-form ideas worth reading from the leading edge of global AI research, products, and industry.
Arvind Narayanan · Akash Kapur · 2026-07-09
Up the Stack: How AI’s Escape From the Commodity Trap Risks Enterprise Lock-in
Arvind Narayanan and Akash Kapur argue that AI labs may escape low-margin model competition by moving up the software stack, while creating stronger switching costs and new risks of enterprise lock-in.
Read article
Jarred Sumner · 2026-07-08
Rewriting Bun in Rust
Jarred Sumner explains why and how Bun’s bundler was rewritten from Zig to Rust, including the agentic coding loops, adversarial review, testing, and performance work behind the migration.
Read article
Anthropic · 2026-07-06
The Making of Claude Code
The researchers, engineers, and early users behind Claude Code recount how an experimental command-line tool grew into an agentic coding product and changed the way they build software.
Read article
Wes Gurnee · Nicholas Sofroniew · Adam Pearce · Mateusz Piotrowski · Isaac Kauvar · Runjin Chen · Anna Soligo · Paul Bogdan · Euan Ong · Rowan Wang · T. Ben Thompson · David Abrahams · Subhash Kantamneni · Emmanuel Ameisen · Joshua Batson · Jack Lindsey · 2026-07-06
Verbalizable Representations Form a Global Workspace in Language Models
Anthropic researchers present evidence that language models develop a small, verbally accessible set of internal representations that acts as a global workspace for report, control, and flexible reasoning.
Read article
Andrew Ng · 2026-07-01
Loop Engineering: Three Key Loops for Building Great Software
Andrew Ng describes three connected loops for building AI-native software: the agentic coding loop, the developer feedback loop, and the external feedback loop.
Read article
Ethan Mollick · 2026-06-30
The Twilight of the Chatbots
Ethan Mollick examines the rapid shift from conversational chatbots to agents that can carry out longer tasks, and what this exponential capability growth means for experts, managers, and institutions.
Read article