Leak claims Anthropic's internal "Model 4" hits ~38-hour task horizon
iruletheworldmo · x · 2026-11-17
Scoop account @iruletheworldmo claims internal Anthropic data is leaking (the "we achieved ASI" framing reads as tongue-in-cheek; unverified).
The attached METR-style chart shows: task length AI can complete at 80% success doubled every 7 months in 2019–2025, accelerating to every 3.3 months in 2024+, and an estimated every 1.7 months since Opus 4.5. Early Claude Mythos Preview measured 3.1 hours, while internal "Model 4" is estimated at 38 hours via the AECI index — a near-vertical curve into 2027. Single-source and partly jokey; awaits official confirmation.
More from Models
- Diffusion images lack controllability: researchers argue for painter-style, tool-call-based generation — josephdviviano · 2026-11-17
- User Has Spent $2,513 on ChatGPT Since 2023 but Eyes a Switch to Claude — MoreLunch1547 · 2026-11-17
- Claude Fable 5.2 writes 43% shorter replies and cuts em dash usage by 99% — ArtificialAnlys · 2026-11-17
- InferenceBench update: Gemini 4 Carbon takes #3, breaking Anthropic's leaderboard sweep — maksym_andr · 2026-11-17
- ChatGPT hallucinates an entire nonexistent client email while sorting the user's inbox — jdjohnson · 2026-11-17
- NVIDIA shares guide to run GLM-5.3 NVFP4 locally on two DGX Stations with vLLM — vllm_project · 2026-11-17