Published a long analysis on open-weights vs proprietary models linkedin.com/posts/federicou…

AI Socratic April Part 2 blog post is ready to read! The topics: 🤖 Models Updates ⏵ @DeepSeek V4 Pro and Flash, 1M token context, Opus 4.7 level and 30x cheaper pricing 💴 ⏵ GPT-5.5 a powerful model — the current consensus is that Codex is better than Claude ⏵ Paper estimates that GPT-5.5 is ~9T params and ⁠⁠Claude Opus 4.7 is ~4T. 💰Fundraising ⏵ Europe: Ineffable Intelligence raises $1.1B seed ⏵ Lidl is the data center of choice in EU 🏛️ Philosophy “If my friend Claudia is not conscious, then what the hell consciousness is?” — @RichardDawkins “Do you know how hard you have to abuse a mammal for them not to have children?” — @NPCollapse Connor Leahy aisocratic.org/blog/ai-socra…

AI Socratic 1.0 in London 🇬🇧 was a success! luma.com/ai-london-1.0 Engineers and founders from Google, AWS, and other local AI startups. We covered Anthropic and Open AI ecosystems, discussed DeepSeek v4, Manus-Meta deal dropping, Cursor-xAI, harness is all you need and more from this blog post: aisocratic.org/blog/ai-socra… Siddhesh Sathe and Aishwarya Sathe presented Tine.sh, a branching notebook runtime for AI and humans. Written in Rust 🦀. @DavideCrapis presentation Scaling Multi-Agent Intelligence is postponed to Shanghai ;) Thanks to connectventures.co for hosting us, @pietrobezza, Rosie, and Claire 🫶 . Big thanks to @saturn_cloud for sponsoring our trip 🙏 See you tomorrow (May 1st) in Lisbon 🇵🇹 👉 luma.com/cicip0jl 👈

first stop London, Apr 28! luma.com/ai-london-1.0

Anissa and I will be traveling in Europe and China to start new AISocratic.org chapters: · London · Lisbon · Berlin · Shanghai In China we'll be inside the Alibaba HQ to meet the top Chinese AI, robotics, EV, and biotech companies. I'm incredibly excited to meet with @Alibaba_Qwen, @BytedanceTalk (SeeDance 2), @Kimi_Moonshot, @Zai_org, @UnitreeRobotics, @Xiaomi, @Midjourney China, and more — and no lies, even more excited to explore museum, activities, and restaurants!!

What a fantastic night @ Frontier Tower in San Francisco 🔥 (sorry for the unflattering picture, I'll post the professional ones after Nvidia GTC) Socratic dialogues and presentations over a Mediterranean dinner 🥗 🧆 with some of the sharpest AI engineers, researchers, and founders from top AI startups in SF. We discussed: - DoD drama:Anthropic drawing a red line with the Pentagon, and its great economic success, while OpenAI loses 100m users — 1% of OpenAI are paying users - The economy of data center moving from mining to AI - War in Middle East escalation will cause AI disruptions — AWS UAE data center bombing, and Strait of Hormuz blockage - The Fuckening - Karpathy's autoresearch - Organoid brains playing Doom and fruit fly brain uploaded into a neural network and into simulations. Then we had an incredible lineup of 8 presentations: — Lance Legel (CEO, Ecological Intelligence): DeepEarth AI Planetary Simulator — Orion Parrott (CTO, OpenFPGA): Fast FPGA LLM token factory for all — Gary Qi (Operation, Trae): Building software in a native way — Deva Temple (Founder & Lead Researcher): Relational Ethics as a Countermeasure to Instrumental Convergence — Junior Rojas (AI Researcher): Morphology-Adaptive Muscle-Driven Locomotion via Attention Mechanisms — Matej Hamas (Senior Tech Lead, Apify): Universal CLI MCP client — Austin Baggio & Sai V. (Ensue): Autoresearch@home — Carlos Lospoy: Model verification in decentralized inference networks Huge thanks to Devinder Sodhi, our new SF ambassador, for co-hosting with me and David Casey (Funding the Commons) for running the 2-days Intelligence at the Frontier conference. Thanks frontiertower for the incredible venue, and Cline for the food. Links: - website: aisocratic.org - presentations: lnkd.in/eCngF_Uz - event: luma.com/ai-sf-2.0 - blog post: lnkd.in/eUXnsWCD Sign up in our website to receive updates on the date of the next AI Socratic SF event ❤️

I just finished writing the AI Socratic March edition. This month topics: - War in Middle East and second order effect on AI - Karpathy autoresearch can train SLM with AI agents - Saaspocalipse & The job market - We put a human organoid brain into Doom and we uploaded a fly brain into a NN Took me a few days to write, so time to take a long walk and touch grass 🌱. Catch you next week in SF and NY or next month somewhere in Europe. aisocratic.org/blog/ai-socra…

I'm searching a venue in NY for the next AI dinner.. who should I talk to? VCs and startups office are cool, but even loft or private space, ideally for 70 guests luma.com/ai-dinner-mar-2026

This was the best AI Socratic event yet, we had 7️⃣0️⃣ top tier guests, tons of great topics and presentations. Sponsors Let me start thanking the sponsors for offering the venue, food, and drinks. • @ditzikow from @spc • @rhobusiness • @meowgorithm from @charmcli Agenda • 🕡 Welcome & dinner 🥗🥙🥤 • 🕢 Socratic Dialogues 🧔🏽‍♂️🏛️ • 🕣 Presentations 👩‍🎤📽️🌌 Socratic Dialogues🧔🏽‍♂️🏛️ (moderated by @feulf) • OpenClaw 🦞 and Moltbook • Opus 4.6, GPT 5.3, Kimi 2.5 • Seedance 2.0 • Geopolitics and decoding "the files" Full list: aisocratic.org/blog/ai-socra… Presentations 👩‍🎤📽️🌌 (moderated by @galic_ivo) • Context graphs for smarter agents (live demo) — Boris Revechkis • Why strong models converge — Jianing Qi • The crypto-agentic runtime layer — @DeleAtanda • Thinking about thinking (humans + LLMs) — Evan Correa • Community Biology × AI — Craig M Trester • AI + religion in Silicon Valley — @calder_mchugh

hey @benln, we're organizing a collaborative Claude Code Workshop on Jan 21st. We'll focus on vibe coding so we'll talk about Cursor too. Please share with your NY team ;) luma.com/ai-dinner-jan-2026?…

bookmark this, one AI leaderboard to track the top benchmarks: aisocratic.org/leaderboard

dear @Google, @OpenAI/@Microsoft, I'm paying an eye for a bootstrap company. The simplest solution for me is to create a new company to get a new round of credit, but that's accounting / bureaucracy hell. I'm building one of the most successful AI community in NYC (lu.ma/ainy) and @FlowAI_xyz a future AI deca-unicorn. Who can I talk to, to receive some extra free credits to keep the light on and to focus on building instead of cutting costs! Thanks, Federico

I just finished updating the AI Socratic Nov blog post site.flowai.xyz/ai-socratic-…

The AI NYC community has a new chapter in Milan! Super excited!

The incredible success of the @cursor_ai got me so inspired! @benln impeccable event. We have an AI happy hour Lu.ma/happy-hour-1.0. Now we need a monthly AI cafe! @galic_ivo @AmShah06 let’s do it. Cafe cursor idea is to meet other people coding. The AI cafe could be a bit broader, so it would need to be very selective to keep a high quality of builders and researchers, and ideally a lightning demo-time at the end, so 3-4 people can present some cool projects. We only need a coffee shop for a full day.

AI Dinner 14.0 at @Sei_Labs 🎃 — here's some great deep tech presentations on Muon, RL + Lora, and OAI Voice API: - Byron Demo, Alon Bochman byron.redshiftlabs.io, an AI sales person that uses the snappy gpt-streaming API - Paper Analysis, Modular Manifolds (Muon) by J. Bernstein: presented by James Smith, Ph.D. and @DotDotJames docs.google.com/presentation… - Teaching LLM to play Wordle with RL and Lora by Charbel K docs.google.com/presentation… And here's the link to my usual presentation on the latest in AI: site.flowai.xyz/ai-socratic-…. Event link: lu.ma/ai-dinner-14.0 Thanks! - @WaverlyMaven and @Sei_Labs for sponsoring the amazing space and drinks - @TrebuchetNet for sponsoring the incredible Indian food - AR Theatrical, vibes and organization - @galic_ivo for helping with the Socratic dialogues moderation See you at the next dinner in November!