--- title: "🔬Scaling Past Informal AI" url: https://daily.dev/posts/scaling-past-informal-ai-9wkniu88p source_url: https://www.latent.space/p/axiom type: article source: "Latent Space" published: 2026-06-03T19:32:18.503Z updated: 2026-06-03T19:32:56.150Z tags: ["reinforcement-learning"] reading_time: 8 upvotes: 0 comments: 0 language: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # 🔬Scaling Past Informal AI **[Latent Space](https://daily.dev/sources/latentspace)** · 8 min read · 0 upvotes · 0 comments ## Summary Axiom, a startup valued at $1.6B, is betting that formal verification using the Lean proof language is the key bottleneck to AGI progress. CEO Carina Hong argues that AI systems relying on informal proofs hit a ceiling, while verified generation enables compounding intelligence — each formally proven result becomes a trusted building block for future training and inference. Axiom achieved 99% (187/189) on the Verina code-with-proof benchmark, compared to OpenAI o3's 4.9%, and solved all 12 Putnam exam problems. The core thesis: formal proofs provide a much stronger RL reward signal than statistical methods like GRPO, improving sample efficiency and enabling a compounding corpus of verified knowledge. Hong argues no path to AGI exists without verified generation. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [When AI Writes the World’s Software, Who Verifies It?](https://daily.dev/posts/when-ai-writes-the-world-s-software-who-verifies-it--tsppkfj9s) · Hacker News · 1 upvotes · 0 comments - [Lean4: How the theorem prover works and why it's the new competitive edge in AI](https://daily.dev/posts/lean4-how-the-theorem-prover-works-and-why-it-s-the-new-competitive-edge-in-ai-wirahwkig) · Hacker News · 2 upvotes · 0 comments --- Tags: [#reinforcement-learning](https://daily.dev/tags/reinforcement-learning) [View this post on daily.dev](https://daily.dev/posts/scaling-past-informal-ai-9wkniu88p)