📊 Source Intelligence
Useful for: Tracking model releases, benchmarks, and how the labs explain themselves.
What we learn from them: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release.
How we compete: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax.
| Source | Group | What it is |
|---|---|---|
| Claude Platform Docs platform.claude.com | Learn & Build | Claude API and Claude Code docs. |
Why it’s on the study list: filed under Learn & Build, Claude API and Claude Code docs. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit platform.claude.com → | ||
| Hugging Face huggingface.co | Learn & Build | Models, datasets, and demos. |
Why it’s on the study list: filed under Learn & Build, Models, datasets, and demos. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit huggingface.co → | ||
| Google AI ai.google | Learn & Build | Research and tools. |
Why it’s on the study list: filed under Learn & Build, Research and tools. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit ai.google → | ||
| OpenAI openai.com | Learn & Build | Models and developer platform. |
Why it’s on the study list: filed under Learn & Build, Models and developer platform. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit openai.com → | ||
| The Batch deeplearning.ai | News & Research | Andrew Ng's AI newsletter. |
Why it’s on the study list: filed under News & Research, Andrew Ng's AI newsletter. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit deeplearning.ai → | ||
| arXiv cs.AI arxiv.org | News & Research | Latest AI papers. |
Why it’s on the study list: filed under News & Research, Latest AI papers. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit arxiv.org → | ||
| Internet Archive archive.org | News & Research | Web and media preservation. |
Why it’s on the study list: filed under News & Research, Web and media preservation. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit archive.org → | ||
| Interconnects interconnects.ai | Ten more worth your time | Nathan Lambert on open models and post-training. |
Why it’s on the study list: filed under Ten more worth your time, Nathan Lambert on open models and post-training. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit interconnects.ai → | ||
| Google DeepMind deepmind.google | Ten more worth your time | Frontier research. |
Why it’s on the study list: filed under Ten more worth your time, Frontier research. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit deepmind.google → | ||
| Stanford HAI hai.stanford.edu | Ten more worth your time | Human-centered AI. |
Why it’s on the study list: filed under Ten more worth your time, Human-centered AI. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit hai.stanford.edu → | ||
| Import AI importai.substack.com | Ten more worth your time | Weekly research digest. |
Why it’s on the study list: filed under Ten more worth your time, Weekly research digest. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit importai.substack.com → | ||
| Hugging Face Papers huggingface.co | Ten more worth your time | Trending papers. |
Why it’s on the study list: filed under Ten more worth your time, Trending papers. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit huggingface.co → | ||
| Transformer Circuits transformer-circuits.pub | Ten more worth your time | Anthropic's interpretability research. |
Why it’s on the study list: filed under Ten more worth your time, Anthropic's interpretability research. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit transformer-circuits.pub → | ||
| Alignment Forum alignmentforum.org | Ten more worth your time | AI-safety research. |
Why it’s on the study list: filed under Ten more worth your time, AI-safety research. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit alignmentforum.org → | ||
| Simon Willison simonwillison.net | Ten more worth your time | Hands-on LLM notes. |
Why it’s on the study list: filed under Ten more worth your time, Hands-on LLM notes. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit simonwillison.net → | ||
| Latent Space latent.space | Ten more worth your time | AI-engineering podcast. |
Why it’s on the study list: filed under Ten more worth your time, AI-engineering podcast. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit latent.space → | ||
| Ahead of AI magazine.sebastianraschka.com | Ten more worth your time | ML deep-dives. |
Why it’s on the study list: filed under Ten more worth your time, ML deep-dives. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit magazine.sebastianraschka.com → | ||
| Anthropic anthropic.com | Labs & Models | Claude models & research. |
Why it’s on the study list: filed under Labs & Models, Claude models & research. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit anthropic.com → | ||
| Meta AI ai.meta.com | Labs & Models | Llama & open research. |
Why it’s on the study list: filed under Labs & Models, Llama & open research. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit ai.meta.com → | ||
| Mistral AI mistral.ai | Labs & Models | Open-weight models. |
Why it’s on the study list: filed under Labs & Models, Open-weight models. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit mistral.ai → | ||
| Ars Technica AI arstechnica.com | News & Learning | Rigorous AI reporting. |
Why it’s on the study list: filed under News & Learning, Rigorous AI reporting. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit arstechnica.com → | ||
| TechCrunch AI techcrunch.com | News & Learning | Startup & funding news. |
Why it’s on the study list: filed under News & Learning, Startup & funding news. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit techcrunch.com → | ||
| fast.ai fast.ai | News & Learning | Practical deep learning. |
Why it’s on the study list: filed under News & Learning, Practical deep learning. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit fast.ai → | ||
| MIT Technology Review technologyreview.com | News & Learning | AI analysis. |
Why it’s on the study list: filed under News & Learning, AI analysis. Study angle: The labs publish dense primary sources; the best aggregators win by translating jargon into plain English within hours of a release. Our edge: We compete on clarity and speed: plain-language guides, updated the day things change, with no hype tax. Visit technologyreview.com → | ||
| Artificial Analysis artificialanalysis.ai | Current · Benchmarks & labs | Independent model benchmarks & pricing |
Why it’s here: current, popular source, Independent model benchmarks & pricing. Visit artificialanalysis.ai → | ||
| LMArena arena.ai | Current · Benchmarks & labs | Crowd-sourced model rankings |
Why it’s here: current, popular source, Crowd-sourced model rankings. Visit arena.ai → | ||
| Epoch AI epoch.ai | Current · Benchmarks & labs | Compute & capability trends |
Why it’s here: current, popular source, Compute & capability trends. Visit epoch.ai → | ||
| The Decoder the-decoder.com | Current · Voices & news | Daily AI news |
Why it’s here: current, popular source, Daily AI news. Visit the-decoder.com → | ||
Who to follow on Claude: the best people and channels on YouTube, X, LinkedIn and Reddit, updated October 25, 2026. See the list on FableGuide →