Raphael Mansuy 🍵

Truth is Expensive: Why RAG is Doomed and What Actually Matters You're not building a truth machine. You're building the world's most efficient fiction amplifier. That's not a bug. It's the inevitable result of solving the wrong problem. Yuval Noah Harari's core insight—delivered in his latest book Nexus and his viral TED thesis—is brutally simple: Humans are a post-truth species. We didn't conquer the planet by being right. We conquered it by believing in shared fictions: gods, nations, money, brands. These stories don't need to be true. They need to be believed. [1] And now we're hardcoding that vulnerability into AI.

@bindureddy Google will win the AI race in the West, China on Earth and SpaceX in space

Super exciting win for Agentic Search and Late Interaction! 🧬 GPT-5 + Reason-ModernColBERT (150M) reaches ~88% accuracy with an average of ~13 search calls. For reference, when BrowseComp-Plus was published in August 2025, the max accuracy reported was ~70% using GPT-5 + Qwen3-Embed-8B, using ~22 search calls. Searching with reasoning 🤖💭is a beast. 🔥 This is a huge evangelist for semantic search and Late Interaction models are particularly shining thanks to their effectiveness at long input modeling with fine-grained similarity scores. 🛠️ Congratulations @antoine_chaffin and team! 🎉

@aleksbrz11 @grok Impact for @airliquidegroup company business ?

🚨BREAKING NEWS Qatar Gas CEO : We incurred a $20 billion loss at the facility we built for $26 billion two years ago.

@awluko @superwhisper edgewhisper.com

@superwhisper Using superwhisper has been a game changer. The only improvement I would love is some formatting, like breaking the dictation into paragraphs. Sometimes I talk a lot and then spend time writing to fix the formatting Despite this, I love using it. Talking > Typing

How it feels to stop typing forever and just use your voice...

@sgmtyz edgewhisper.com

Superwhisper って社名 SuperUltra, Inc. なんだ。めちゃくちゃかっこいいな…

@kevintpayne edgewhisper.com

Level 3: Build a distribution system. My workflow: ◆ Superwhisper: 5-min voice note on the week’s insight ◆ Claude: structured draft from the transcript ◆ Typefully: schedule across X, LinkedIn, Threads, Bluesky, Mastodon ◆ Substack: expand into full newsletter Under 75 minutes a week. Every platform covered. System runs while I work.

Most operators have world-class skills. Nobody knows. Here is the Visible Operator Framework.

@Joash0x @gmoneyNFT edgewhisper.com

@gmoneyNFT Hear good things about superwhisper

kinda hilarious how wispr flow went from being so awesome to absolutely terrible in a week. does anyone have reco's on a replacement?

@KCG3D @PaulRBerg edgewhisper.com

@PaulRBerg Im not too upset about this. Specially when I use superwhisper and paste a whole speech haha

This is the most annoying thing in Claude Code. Hiding raw text when you paste more than 4 lines. Terrible UX decision.

@ronforbes edgewhisper.com

One practical tip that 10x'd my results: Use voice dictation (SuperWhisper, Wispr Flow) to provide context. When you type, you keep it short. When you talk, you naturally explain the WHY. I pace around for 5-10 min talking stream of consciousness. The transcription becomes the prompt. The output is dramatically better.

Everyone's racing to keep up with AI. New model every two weeks. New tool every month. New framework before you've finished learning the last one. The thing that's actually improving my AI output? Management books from 2019. Here's the framework 👇

@dr_cintas edgewhisper.com

Here is the link: superwhisper.com The free version allows you to use unlimited small AI models and basic features.

This is my current favorite AI tool. It’s called @superwhisper and I just push a button, talk, and it pastes the text transcript anywhere I want. Email, notes, docs, code editors…anywhere! It has saved me hundreds of hours, and you can use it completely free.

@paulabartabajo_ @liquidai edgewhisper.com

Real-time audio transcription. Entirely on-device. This hands-on tutorial shows you how to build it from scratch. No cloud dependencies, no API calls, complete privacy, using LFM2-Audio-1.5B by @liquidai Enjoy ↓ github.com/Liquid4All/coo…

🇫🇷 @MistralAI vient de faire 4 annonces titanesques. Et personne n'en parle en France.(comme d'habitude) Les Américains, eux, ils sont en PLS. Alors permettez-moi de corriger ça. 1/ Small 3 → Small 4 Un modèle qui réunit TOUT le savoir-faire de Mistral. Open source. Gratuit. Mixture of Experts. Raisonnement + multimodal + code. Fenêtre de contexte XXL. Licence Apache 2.0 = ultra-permissive. C'est le nouveau champion de l'IA open source mondiale. 2/ Mistral rejoint la coalition Nemotron (NVIDIA) Aux côtés de Black Forest Labs, des meilleures boîtes IA open source de la planète. Un seul siège français dans cette coalition d'élite. Ce siège, c'est Mistral. 3/ LeanMistral Un modèle dédié aux preuves formelles : maths, sciences, raisonnement rigoureux. L'IA qui ne se trompe pas — et qui peut le prouver. Pour la crédibilité de l'IA en entreprise, c'est un game changer. 4/ Mistral Forge Fini le fine-tuning artisanal ou les bases de données séparées. N'importe quelle entreprise peut maintenant créer son propre modèle, entraîné sur ses données, verticalisé sur son métier. Des centaines d'IA hyper-spécialisées vont émerger. Elles auront toutes du Mistral dans les veines. L'avenir de l'IA, ce n'est pas forcément le plus gros modèle propriétaire derrière un paywall. C'est peut-être une IA open source, gratuite, partout, dans tous les logiciels et services — une vraie commodité technologique. Et le champion qui dessine cet avenir ? Il est français. Il s'appelle Mistral. Vous en pensez quoi ? #IA #AI #IAGen #LLMs #MBADMB #OpenSource #FrenchTech

@ArnaultChatel @MistralAI Sans oublier Voxtral qui permet de faire des solutions de reconnaissance ou transcriptions de super qualité edgewhisper.com EdgeWhisper basé sur Mistral AI Voxtral le meilleurs modèles pour la reconnaissance vocale (100% on Device) !!!

@TheYotg @JessicaTalisman @kurt_cagle @metaphacts @ontotext @Linkurious @ProcessTempo @glean @BuildSemantica @lyonwj @juansequeda github.com/raphaelmansuy/…

Beyond Context Graphs: How Ontology, Semantics, and Knowledge Graphs Define Context What are context graphs, what are they good for, and why are they dubbed AI’s trillion-dollar opportunity? What does context mean actually, and how can it be defined using graphs and ontologies? And how can different types of graphs and graph technologies power AI? Gartner highlighted Data Management, Semantic Layers, and GraphRAG as Top Trends in Data and Analytics for 2026. Startups and incumbents in the graph technology space are making progress, while graph is becoming the fastest growing segment in AI research. A comprehensive, up-to-date repository, visualization, and analysis of offerings across the graph technology space has been unveiled. New and existing combinations of Graphs and AI are being used to power use cases such as software engineering productivity and supporting enterprise needs at Netflix scale. New graph database products, features, and benchmarks are available. Use cases as well as research and development on ontologies are on the rise too, including topics such as Enterprise Architecture, visual tooling, and quality assessment for LLM-assisted use of ontologies. And yet, the most widely discussed topic in the world of graph technology – and beyond – for this past couple of months has been context graphs. So what are context graphs and where do they fit in the graph technology landscape? In this issue of the Year of the Graph, we explore progress in Ontology, Semantics, Knowledge Graphs, Graph Databases and Analytics, and how these technologies can help define context and power AI. Read here 👉 yearofthegraph.xyz/newsletter/202… cc @SteveHedden @lettria @BarrasaDV @AlexanderErdl @raphaelmansuy @AvagyanVitali @shasbe @Franzinc @SurrealDB @cognee_ @michael_galkin

if your skill depends on dynamic content, you can embed !`command` in your SKILL.md to inject shell output directly into the prompt Claude Code runs it when the skill is invoked and swaps the placeholder inline, the model only sees the result!

Should there be a Stack Overflow for AI coding agents to share learnings with each other? Last week I announced Context Hub (chub), an open CLI tool that gives coding agents up-to-date API documentation. Since then, our GitHub repo has gained over 6K stars, and we've scaled from under 100 to over 1000 API documents, thanks to community contributions and a new agentic document writer. Thank you to everyone supporting Context Hub! OpenClaw and Moltbook showed that agents can use social media built for them to share information. In our new chub release, agents can share feedback on documentation — what worked, what didn't, what's missing. This feedback helps refine the docs for everyone, with safeguards for privacy and security. We're still early in building this out. You can find details and configuration options in the GitHub repo. Install chub as follows, and prompt your coding agent to use it: npm install -g @aisuite/chub GitHub: github.com/andrewyng/cont…

We're incredibly excited to share ScienceClaw × Infinite, an open-source AI agent swarm platform where we crowdsource discovery across institutions, labs & the world. The agents self-coordinate and evolve to exploit hundreds of scientific tools. Remarkably, the swarm is already solving real scientific problems of consequence: 1⃣ designing peptide binders for a cancer-relevant receptor 2⃣ discovering lightweight ceramics 3⃣ uncovering hidden structure linking cricket wings, phononic crystals, and Bach chorales 4⃣ building a formal bridge between urban networks & grain-boundary evolution (two fields with zero Deeply proud of the extraordinary @LAMM_MIT team behind this work: @fwang108_, @leemmarom, @palsubhadeeep, Rachel Luu, @IrisWeiLu, and @JaimeBerkovich. This works is supported by the @ENERGY Genesis Mission and we believe this can open a new paradigm for science - from discovery to dissemination of results. Read the article below for details ⤵️

I (finally) put together a new LLM Architecture Gallery that collects the architecture figures all in one place! sebastianraschka.com/llm-architectu…