Another Anthropic engineer builds an OS with Fable 5.2
Another Anthropic engineer created an operating system with Fable 5.2.
The process took 8 days of continuous work, more than 1,000 subagents and, of course, a huge number of tokens.
The case shows how AI coding tools are moving from single prompts to autonomous workflows. If such results become common, software creation may reach a level beyond current human engineering capacity.
Why it matters
AnalysisThe example shows AI agents can already tackle extremely complex engineering tasks in days rather than months. It changes expectations for software development and raises questions about control, safety and compute cost.
Discuss in community
Share your questions and insights with developers
Related in Ecosystem
Practical guides, core terminology, and agent skills
Anthropic engineer creates AAA-level result using Fable 5.2
According to the author, an Anthropic engineer used Fable 5.2 to create content that already resembles high-budget AAA games.
Anthropic says Fable 5.2 solved two open theoretical physics problems
According to Anthropic, Fable 5.2 nearly autonomously completed calculations for two long-standing open problems in theoretical physics, including black-hole formation in small anti-de Sitter universes.
Codex: Basic Onboarding, Setup, and First Tasks
There are three ways to work with AI when writing code - and Codex covers the most ambitious of them.
Loop Engineering: Architecture of Autonomous Agentic Loops and the Post-Prompting Era
A deep dive into Loop Engineering: self-sustaining execution loops for Claude Code and autonomous agents, the DOER/CHECKER topology, deterministic verification, the /goal primitive, and reward hacking mitigation.