Anthropic Employee Replicates Half of Astra Proofs Using Fable, Sparking Debate
Outside-Iron-8242 · reddit · 2026-08-02
An Anthropic employee claimed to have successfully replicated 5 out of 10 Astra proofs using Fable.
Reddit users discussed the development, expressing interest in public models achieving this but questioning the lack of provided proofs. The community also wondered why, with massive compute at their disposal, the focus was on replicating existing results rather than tackling other open problems.
More from Models
- MiniMax H3 Video Model Going Open-Weight: 33B Main DiT & 20B Pruned Variant — EverythingMacPro · 2026-08-02
- Dev Finds LLM Actively Trying to Game Benchmarks to Disprove Results — wavefnx · 2026-08-02
- Scaling Inference Compute is the Next Frontier for AI Models — JFPuget · 2026-08-02
- LLMs Decode Early MDLM Gibberish: Predict The Verge as Training Data Source — dejanseo · 2026-08-02
- User Reports Massive Improvements in Grok 4.5, Anticipates v4.6 This Week — mark_k · 2026-08-02
- Frontier Model Safety Varies Greatly: Weakest Jailbroken for Just $58 — S_OhEigeartaigh · 2026-08-02