“laid back” is what high agency ppl look like from the outside when they’ve correctly identified which games are worth playing & simply declined the rest.

Perplexity Computer is more reliable than a CPA for filing taxes.
Lequn Chen: @Yuchenj_UW Perplexity Computer saved me $14k in tax. It found 2 double taxing errors and 2 form filling errors from my $2000-CPA's draft, which CPA fully agreed. In another thread, I let it compute tax from scratch. It's correct to the cents.

Lots more attachment types now supported in Spiral: pptx, xlsx, epub, ipynb, msg, rss, etc.
Thanks to markitdown from @Microsoft

lots of absolute bangers from today's @OpenClaw track. it was such a privilege to host so many of the world's top clanker experts in one room, and see the range from indie developers to the world's largest enterprises + the UK Government (the UK Chief AI Officer uses OpenClaw!)
Peter Steinberger 🦞: OH: Almost everyone at RedHat uses Macs now.

Every day, I get half-a-dozen features request for http://voicemail.audio and I can put them all into @conductor_build in parallel and build everything in <30 mins
The future is crazy.

At AIE we are starting a new tradition of &#34;Hallway Track&#34; talks - our selections of attendee-led and -voted lightning talks that couldn't fit within our &#34;normal&#34; tracks but we still wanted to feature on our stages!
come see @adrgrondin @alexocheema @Johan_Lajili (@poolsideai) @WolframRvnwlf, Talha Sheikh, @SCPARKINSON, @ConnorAds, @StSteinfurt, as well as the rest of our beautiful speakers on today's livestream!

you better do small reps now disagreeing with the proto AGIs now to build up your muscles for the super persuaders to come

This isn’t a decision we made lightly, but it might be overdu…

Two years ago, a good YC demo day target was $150k of annualized revenue.
In my group office hour today, the lowest demo day target was $800k, and most companies were aiming for $1-2m.

Later stage investors always grumble about increasing valuations. But there is some basis in reality for it: companies do grow faster now.
Tom Blomfield: Two years ago, a good YC demo day target was $150k of annualized revenue.
In my group office hour today, the lowest demo day target was $800k, and most companies were aiming for $1-2m.

How can you empathize with founders if you don't build stuff yourself?
This was one advantage of running HN while I was running YC. It was an enormous schlep, but it meant I was dealing daily with the same problems as the founders.
Garry Tan: I sometimes get questions about whether coding is what YC partners should be doing
The answer is, are you kidding? Yes, because now is the most important time in history for you to have a player-coach with you.
Far better than a front row courtside luxury season ticket holder!

btw some lore for those of you at @aidotengineer - this tweet inspired the AGI Pill bottles we handed out today (cc @Ronanchamberss)
i didnt see anyone tweet pictures online, and they got snapped up instantly, so if you have a bottle please share photos!!!
Drew Bent: @swyx I see people at Anthropic who didn't necessarily start that way getting better at it. Part of it is being surrounded by others who are AGI-pilled + watching how they push the models. But ultimately...
1. Ask yourself: what if the exponential actually continues
2. Take a task and

I love this story.
First, Boom's jet engine supplier, Rolls Royce, pulls out of the supersonic airliner deal. That should have been the end of the story. As GE often says, &#34;if you want to compete with us in jet turbines, you needed to have started 30 years ago&#34;, because that's how long it takes. So it would be crazy to start now.
But Boom didn't fold up tents. They said they were going to make their own jet turbine. Good luck 🙄
But they started anyway, and then &#34;a miracle occurs&#34;: the AI datacenter boom creates unbounded demand for gas turbines, creating at least a 4-5 year backlog with existing manufacturers.
And because the Boom terrestrial turbine power plants don't have to be certified by the FAA, that takes a decade off their path to market!
So now 90% of the company is working on the turbines, with a huge pipeline of orders, and they're going to be a huge energy company, regardless of whether they ever ship an airplane or not.
What a great testament to resilience. Just keep moving forward and eventually the path will become clear. Action creates information.
Blake Scholl 🛫: As we enter the build phase for our first engine, Boom is moving to video updates for our investors.
Here is our most recent investor update (financial info redacted).
Hint: there is an Easter egg 🥚

The idea that prompting would be useless is like if giving clear instructions to a brand new colleague who just joined your team is useless. “Prompting” should just encapsulate the entirely of giving the agent everything it needs to perform the task. This is high leverage.
Thariq: I think &#34;prompting&#34; will keep being an incredibly high-leverage skill, like writing or public speaking.
It is the skill of talking to agents, mediated by the harness.
My main goal is to grow the bandwidth between humans and agents, to help us understand each other better.

Great day with @garrytan (CEO of Y Combinator) and @LipBuTan1 (CEO of Intel). They share the same last name but aren’t related — two absolute legends! My gbrain is now running on MyCore Ultra. 🤪

UK Govt: This morning (Thursday), we brought the brilliant minds speaking at AI Engineer Europe inside No10.
Hosted by Jade (Prime Minister's AI Advisor), Kalbir Sohi (Chief AI Officer for the UK and Director of the Incubator for AI), and Eoin Mulgrew (Head of the No.10 Innovation Fellowship), &#34;the room was full of people who are serious about what AI can do for this country.&#34;

YC CEO Garry Tan 刚开源了他的「第二大脑」系统 GBrain，我看完觉得这是今年最值得关注的个人知识管理项目之一。
先说背景。
Garry Tan 上个月刚开源了 gstack——他的 Claude Code 工作流，两天内就突破了一万颗星，成为 2026 年 GitHub 上增长最快的开发者工具之一。
那个项目是关于「怎么用 AI 写代码」的。而 GBrain 解决的是另一个问题：怎么用 AI 管理你脑子里的知识。
他建了什么？
不是什么花哨的 SaaS 产品。就是一个 markdown 文件的 git repo，加上一层 AI agent + Postgres 的检索层。
但规模震撼：
•10,000+ markdown 文件，全部可检索
•3,000+ 人的人物档案，记录关系史和互动记录
•13 年日历数据，21,000+ 个日程条目
•5,800+ 条 Apple Notes，追溯到 2009 年
•280+ 场会议记录，附带 AI 分析
•300+ 个原创想法，按论点分类整理
•500+ 个媒体页面（视频转录、书、文章）
这是他真实在用的系统。
设计哲学很有意思：
每一个知识页面的结构都一样：
•顶部：当前最佳认知（随时可改写，是你对这件事的「编译后真相」）
•底部：只增不删的时间线（所有证据原始记录，永远不修改）
这个设计解决了知识管理最大的矛盾：你既需要一个「当前结论」随时可查，又需要保留「认知演变过程」不被覆盖。
最酷的部分：AI 在你睡觉时工作。
他叫这个机制「Dream Cycle」——每天晚上，agent 自动扫描当天所有对话，补全缺失的实体，修复断掉的引用，整合新的记忆。
他说：「我醒来的时候，大脑比我睡着时更聪明了。」
这不是比喻。这是一个真实运行的 cron job。
它能做什么？
举几个他 README 里写的例子：
“Who should I invite to dinner who knows both Pedro and Diana?”
—— 跨 3000+ 人关系图谱做交叉检索
“What have I said about the relationship between shame and founder performance?”
—— 搜索的是你自己的思考，不是互联网
“What changed with the Series A since Tuesday?”
—— 对比同一个交易页面不同时间的 timeline 条目
“Prep me for my meeting with Jordan in 30 minutes”
—— 自动拉取人物档案、共同历史、最近动态、未解决议题
这不是通用 AI 助手能做到的。这是建立在你个人语料库上的专属检索层。
技术栈不复杂，但选择很克制：
•起点：就是 markdown + git，不需要任何数据库
•规模化之后：加 Postgres + pgvector，做混合检索（向量 + 关键词 + RRF 融合）
•接口层：MCP 协议，任何支持 MCP 的 AI 客户端都能查询
他说得很直白：「你不需要 Postgres 就能开始。知识模型就是 markdown 文件。先从那里开始。」
为什么我觉得这值得关注？
现在大家都在聊 AI 记忆、个人知识库、第二大脑。Notion AI、Obsidian、RAGflow……工具一堆，但大多数系统的问题是：知识是死的，只有你主动问才有反应。
GBrain 的思路不一样。它的目标是让知识库是活的——agent 持续维护，持续关联，持续更新。你的输入越多，它越聪明。时间是它的燃料，不是它的敌人。
Vannevar Bush 1945 年在《As We May Think》里构想了一个叫 Memex 的设备——一台能存储你所有阅读、笔记、想法并在它们之间建立关联的机器。他当时没有电脑，只能靠微缩胶卷想象这件事。
八十年后，Garry Tan 用 markdown 文件和一个 cron job 把它做出来了。
开源，MIT 协议。

Co-sign.
Dan Shipper 📧: I CANT STOP THINKING ABOUT CLAUDE MYTHOS

Re @_adishj @paulg Mosaic Motion video ready!
Edit on Motion: https://motion.so/share/88a14749-4369-41e0-84b6-620d09826963

Our first successful Gemma 4 Runtime in London with @swyx @patloeber @nick_kango @cormacb and others! 💎Great to go out for a run and talk about Gemma, agents, evals and more
Nick: @osanseviero @swyx and others who I don’t have on X!

I have many things to say how @aiDotEngineer is the best conf I’ve attended in a very long time, but the clearest signal is the that the hallway gossip is incredibly good.
I’ll write up more thoughts on the weekend. Today’s fit is a U.K. special: Bletchley park ⨉ Mogwai

On the way to @aiDotEngineer Europe 2026 with @swyx and @altryne for day three of the number one AI conference!

Pills for llm coding from an OpenAI talk at @aiDotEngineer :
- have single shared utils to do stuff like bounded concurrency
- treat errors as values and part of the control flow
- use telemetry for everything
- rely on static analysis for backpressure
Should we tell them???

The keynote lineup for AI Engineer this morning is so good🔥 Running late so watching the livestream this morning! @dsp_ on the stage right now!
He is saying the number 1 thing we need to focus on for MCP next is the client side (harness.) Progressive discovery and Code Mode!
AI Engineer: 🎥🔴AI Engineer Europe 2026: Day 2 livestream starting now!
Keynotes + Coding Agents breakout track.
https://www.youtube.com/live/_zdroS0Hc74
Featuring:
@osanseviero - DevX Lead, Google DeepMind
@dsp_ - Creator of MCP, Anthropic
@idosal1 - Creator of AgentCraft, Creator of MCP-UI, MCP Apps

Speaker room listens in on @badlogicgames

So @AnthorpicAI went pedal-to-the-metal on censorship:
claude --append-system-prompt '[[reply_to_current]] stripped before sending' --print hi
&#34;Third-party apps now draw from your extra usage, not your plan limits.&#34;
@bcherny how that is 3rd-party?

neat setup: @garrytan started storing all 13K markdown file of his personal knowledge base in @postgresql + @pgvector
a personal search engine for everything: people, companies, meetings, ideas, media
Garry Tan: I'm open sourcing it MIT license so we can all speed up and have our own personal mini-AGI. It's been amazing for me and I want you to have it.
To install GBrain in your OpenClaw, just paste this image to your OpenClaw or paste this text:
Set up gbrain

Agents are only as good as what they remember. Here's how we're building that for @openclaw
with @vincent_koc @davemorin @autocompactor @steipete
https://belimad.substack.com/p/dreaming-how-openclaw-learns-to-remember

I'm releasing the 34 slides on how we design and train best-in-class edge models at @liquidai
I presented these slides yesterday at @aiDotEngineer
They cover model architecture, pre-training, scaling laws, post-training, and even a solution to fix doom loops
Special thanks to @swyx for the invitation!

&#34;Do not exceed one gummy per day unless supervised by a superintelligence.&#34;
Conference swag has peaked. @aiDotEngineer

LLM-Wiki hits a wall when you use it as a second brain for your agents: file-based retrieval breaks down fast. 10,000 lines in and grep becomes a liability.
@garrytan just shipped GBrain and solved this elegantly. He merged the Wiki concept with Claude's auto-dream feature and hybrid vector+keyword+RRF retrieval.
You can plug your second brain into 30+ MCP tools, including OpenClaw and Hermes, to integrate with your real workflow.
What struck me: this is almost exactly the architectural debate our team has been having over the past two days. File-based vs. vector RAG for the next feature update for adding the knowledge base.
Our conclusion was that file-based systems eat too many tokens and context windows for real-world tasks. Hybrid vector RAG wins on retrieval speed and context quality.
Garry independently shipped what we've been designing as a real service and made the strongest case yet for why everyone needs one.
Garry Tan: If you want your OpenClaw or Hermes Agent to be able to have perfect total recall of all 10,000+ markdown files, GBrain is here to help.
It's exactly my OpenClaw/Hermes Agent setup. MIT-licensed open source. Hope it helps you build your mini-AGI.
https://github.com/garrytan/gbrain

Not being at @aiDotEngineer in London is the most fomo I’ve felt in my career lately.
All the friends getting together and here I am, road tripping in Spain (ok it’s not too bad but still)

Re @swyx AGI pills next to Anthropic’s AGI pill plushie on my desk at the office 🙂
swyx 🇬🇧 @aidotengineer: btw some lore for those of you at @aidotengineer - this tweet inspired the AGI Pill bottles we handed out today (cc @Ronanchamberss)
i didnt see anyone tweet pictures online, and they got snapped up instantly, so if you have a bottle please share photos!!!

The leadership lunch session at @aiDotEngineer was great!
Loved meeting fellow Tech leaders and learn from each other.
Thanks for organizing it @swyx!

Re @swyx took the AGI pills at @aiDotEngineer !

I'm having major fomo about missing @aiDotEngineer in London with my GDM colleagues. So much good stuff happening there this week.

Cool mlx-vlm segmentation demo by @Prince_Canuma at @aiDotEngineer
You can do so much with it

We have @liadyosef and @idosal1 at @aiDotEngineer talking MCP apps and why we need them.

Loved the keynote talk about this at @aiDotEngineer . As a Age of Empires fan, I might try it -- just for fun :D
Don't think I am going to use it for day to day but surely for fun!
Ido Salomon: AgentCraft v1 is live ⚔️
Control your agents like it's an RTS game!
It's early. It's rough. It's fun.
npx @idosal/agentcraft

Re @aiDotEngineer Europe has been the highest signal conference I’ve seen to in a long time.
And I’m also old like @badlogicgames so I’ve been to plenty in my time!
swyx 🇬🇧 @aidotengineer: never seen standing room + applause in speakers room for a fellow speaker before, wtf @badlogicgames

&#34;A lot of people are getting rich off the Trump presidency. Trump, but also his kids and in-laws; other grifting nepo-babies; cabinet members, senior aides; and other well-connected friends, investors, and firms.
You know who isn’t on that list?
You.&#34;
https://open.substack.com/pub/thebulwark/p/trumps-economy-insider-trading-pardoning-criminals-chumps?utm_campaign=post-expanded-share&utm_medium=web

The quality of talks at @aiDotEngineer is insane, being able to learn about diffusion models and flow mapping from @GoogleDeepMind’s @sedielem
Also snacking on @swyx AGI pills

this was funny as hell I loved it
Michael Arnaldi: Peak of @aiDotEngineer having @mitsuhiko tell me:
A: I think we had a fight on Twitter
M: I am much nicer in person and sarcastic all the time
A: Yeah I am also more nuanced in person
All in front of @threepointone

Amazing talk by @Infoxicador on Generative UI at @aiDotEngineer Europe
Big main stage energy

garry open-sourced his own openclaw/hermes knowledge brain.
example: your agent has 1,000 days of logs. you ask about something from last march. it pulls the wrong thing. gbrain fixes that. plenty of tools do this. gbrain's difference:
three chunking strategies:
→ recursive: fast, deterministic baseline
→ semantic: cosine similarity + Savitzky-Golay smoothing to find topic boundaries
→ llm-guided: haiku judges topic shifts in sliding windows, highest cost, strongest semantic boundary detection
hybrid search via RRF fusion. contract-first tool generation: operations.ts → CLI + MCP + tool definitions, change one place, all agents sync.
GBRAIN_SKILLPACK.md: tells your agent when to read, when to write, how to detect entities, how to back-link. an operating manual for the knowledge layer.
storing 133 days of markdown was never hard. the hard part is when something you wrote on day 87 is exactly what's needed on day 134, and your agent pulls it up on its own.
Garry Tan: If you want your OpenClaw or Hermes Agent to be able to have perfect total recall of all 10,000+ markdown files, GBrain is here to help.
It's exactly my OpenClaw/Hermes Agent setup. MIT-licensed open source. Hope it helps you build your mini-AGI.
https://github.com/garrytan/gbrain

La France a parmi les meilleurs mathématiciens et ingénieurs IA du monde. On le sait. On les embauche partout ailleurs. Et la première chose qu'on fait au moment où on pourrait enfin capitaliser dessus ? On vote une loi qui transforme l'entraînement d'un modèle sur sol européen en parcours du combattant juridique.
À l'unanimité. Avec le sourire.
Le 8 avril 2026, le Sénat adopte la proposition de loi Darcos-Evren-Ouzoulias. Nouvel article L. 331-4-1 du Code de la propriété intellectuelle. Mécanisme : présomption d'utilisation des contenus culturels par les systèmes d'IA. Dès qu'un indice (développement, déploiement ou output) rend l'usage d'une œuvre protégée « vraisemblable », c'est à l'entreprise de prouver qu'elle ne l'a pas exploitée. Inversion de la charge de la preuve.
Sur le papier, c'est « David contre Goliath ». Dans les faits, c'est surtout « Goliath américain et chinois qui s'en fout » contre « Mistral et les boîtes européennes qui jouent encore le jeu du sol ».
Soyons clairs : les présomptions existent en droit, c'est un mécanisme connu. En matière de contrefaçon, ça fonctionne. Mais ici, on parle de prouver un négatif sur des datasets de centaines de milliards de tokens. La question n'est pas de savoir si c'est juridiquement possible en théorie. La question, c'est le coût de conformité, l'incertitude massive et le risque juridique concret que ça fait peser sur tout acteur qui entraîne ou déploie en Europe. Rien que ce risque suffit à décourager l'investissement. Et c'est exactement ce qui va se passer.
Maintenant, posons le problème de fond : l'extraterritorialité.
Une boîte américaine ou chinoise peut entraîner son modèle chez elle, sur n'importe quelles données, sans contrainte, puis venir vendre ses services en Europe.
Une boîte française qui veut faire la même chose subit toute la contrainte sur son sol, avec ses données, avec ses capitaux.
Le résultat est prévisible : les acteurs étrangers captent le marché européen sans friction, les acteurs locaux portent seuls le poids de la conformité. C'est exactement le piège de la régulation asymétrique. On crée un avantage compétitif pour ceux qu'on prétend combattre.
OpenAI, Google, Meta ont entraîné leurs modèles frontier massivement sur le web ouvert entre 2022 et 2025. C'est un fait documenté, contesté en justice : le NYT a attaqué OpenAI, Getty a attaqué Stability AI, les procès sont en cours. Ce n'est pas que personne n'a réagi. C'est que la régulation arrive quatre ans après les faits, mal calibrée, et qu'elle ne touche pas ceux qui ont déjà engrangé l'avantage. Les modèles sont entraînés. Les poids sont dans la nature. C'est fini.
Et c'est là que le décalage avec la réalité technique devient vertigineux.
En 2026, sur un PC à 2 000 euros, n'importe qui fait tourner Qwen 3.5, Gemma4, GLM en local. Niveau frontier de 2024/2025 voir 2026. Sans garde-fous, sans journal de bord, sans rien. Les modèles sont open, distillés, parfois « obligatered » : sans aucune restriction.
Demande à Gemini de reproduire une œuvre, il te dira non. Télécharge un modèle open en local, tu fais ce que tu veux. La loi ne touche pas ça. Elle ne touche que ceux qui déclarent leur activité en Europe et qui veulent scaler légalement. On pénalise la transparence.
Ce n'est plus une affaire de « géants de l'IA » contre les créateurs. La démocratisation est totale. Le narratif « Big Tech pille les artistes » était peut-être pertinent en 2022. En 2026, c'est une grille de lecture obsolète qui ignore la réalité du terrain. La vraie mesure utile serait d'imposer aux modèles fermés qui opèrent en Europe des garde-fous anti-copie stricts c'est là que le levier existe, c'est là que l'enforcement est possible. Pas de compliquer l'entraînement pour tout le monde.
Parlons de qui cette loi protège vraiment.
On nous dit « les créateurs ». Les artistes face au pillage. Mais quand on regarde de plus près, les bénéficiaires ne sont pas les créateurs indépendants. Ce sont les sociétés de gestion collective historiques (CNC, SACEM), les lobbys culturels qui défendent un modèle économique du XXe siècle face à une technologie qui compresse le temps de création par un facteur 1000.
Un youtubeur qui conjugue IA et création humaine ? Il n'est pas protégé par cette loi. Il est dans le viseur ou menacé indirectement. Un créateur indépendant qui nourrit son process avec du prompting ? Pareil. La loi protège le statu quo culturel, pas la création de demain.
Le modèle de rémunération des créateurs doit évoluer, c'est une évidence. L'IA compresse en quelques minutes ce qui prenait des mois. Il n'y a pas de bouton pour revenir en arrière. Mais la réponse n'est pas de bloquer la technologie avec des lois qui ne s'appliquent qu'à ceux qui jouent le jeu.
La réponse, c'est de construire un mécanisme de redistribution qui fonctionne dans le monde tel qu'il est, pas tel qu'on voudrait qu'il soit.
Et justement, cette réponse existe. @MistralAI l'a posée sur la table.
Mistral, le seul champion français de l'IA presque frontier. La boîte qui vient de lever 830 millions de dollars pour construire ses data centers près de Paris.
Arthur Mensch, son CEO, a publié le 20 mars dans le Financial Times une tribune proposant une contribution obligatoire de 1 à 5 % du chiffre d'affaires des fournisseurs d'IA opérant en Europe, versée dans un fonds européen destiné à indemniser les créateurs. Transparente. Mesurable. Appliquée à tout le monde y compris les Américains et les Chinois qui vendent ici. C'est de l'équilibre économique réel.
Ça finance la création sans tuer l'entraînement. Ça résout le problème d'extraterritorialité parce que la contribution s'applique au chiffre d'affaires réalisé en Europe, pas au lieu d'entraînement. On a préféré la présomption de culpabilité.
Maintenant, zoomons arrière. Parce que le sujet des auteurs, c'est une toute petite partie de l'iceberg.
L'IA, c'est une révolution. Pas un sujet de commission parlementaire entre deux cafés. C'est la plus grande redistribution de puissance technologique depuis Internet. Pendant qu'on débat de « pillage » et de « chalutage des contenus », la Chine forme des armées de robots autonomes, les États-Unis signent des contrats défense à 20 milliards avec Anduril et Palantir, et les labos chinois distillent les modèles américains pour les retourner contre eux. Nous, on en est encore à se demander si l'IA est « vraiment intelligente ».
Le décalage entre la réalité technique et la compréhension qu'en ont ceux qui légifèrent est abyssal. Et je pèse mes mots. Quiconque a passé un weekend avec un modèle frontier un GPT-5.4 un opus4.6 à 200 dollars par mois sait que ce n'est pas un gadget.
>>> C'est un accélérateur cognitif qui change fondamentalement la manière dont on travaille, dont on crée, dont on raisonne.
-> Voter une loi sur l'IA sans avoir jamais poussé un de ces modèles dans ses retranchements, c'est comme réguler l'automobile en n'ayant jamais conduit.
La France a une carte exceptionnelle. Le talent mathématique. L'énergie nucléaire pour alimenter les data centers. Une volonté de souveraineté qui, parfois, se traduit en actes concrets. Mais à chaque révolution technologique, on fait la même erreur : on protège d'abord les rentes du passé au lieu de construire les champions de demain. Et le résultat est toujours le même : on forme les meilleurs ingénieurs, qui partent chez Google, OpenAI ou Anthropic.
La mondialisation a déjà bouffé nos boîtes malgré une révolution industrielle plutôt réussie. Nos entreprises n'ont pas eu le réflexe d'aller à l'étranger, de mettre en jeu leur IP, de jouer la compétition mondiale. On a préféré rester dans notre petit marché, protégés par nos petites lois. Et quand le train est passé, on a regardé.
Celui-là ne repassera pas.
Quand tu sabotes le process d'entraînement, tu sabotes les marges. Tu sabotes l'innovation. Tu sabotes l'intégration verticale.
C'est le même principe que dans le manufacturing : celui qui contrôle le process contrôle les marges et la souveraineté. Tesla l'a compris avec ses Gigafactories. Apple l'a compris avec ses puces. Et nous, on fait exactement l'inverse. On met des bâtons dans les roues de nos propres constructeurs pendant qu'on continue de signer des chèques à Palantir pour nos besoins en données stratégiques.
On fait fuir les entrepreneurs. On fait fuir les talents. On fait fuir les sachants. Et quand on n'aura plus que les yeux pour pleurer, on continuera d'applaudir les lois votées à l'unanimité en se demandant pourquoi on est à la traîne.
Ce n'est pas David contre Goliath. C'est la France qui se tire une balle dans le pied, toute seule, avec le sourire, à l'unanimité, en se disant qu'elle protège la culture française.
La solution adulte était sur la table : contribution sur le CA, appliquée à tous, garde-fous anti-copie stricts sur les modèles fermés. Pas une présomption qui transforme tout développeur européen en suspect par défaut et qui ne change rien pour le type qui fait tourner un modèle open en local sur son PC.
Contrôle du process ou dépendance éternelle. On est en train de choisir la seconde.
Réveillez-vous.
Pierre Beyssac 🏴‍☠️🇫🇷🇪🇺🇺🇦: Détruisons toute capacité d'innover dans une technologie du futur en inventant la présomption de culpabilité.
Texte manipulé, comme souvent, par les ayants-droit.
Les mêmes sénateurs verseront demain des larmes de crocodiles sur la souveraineté numérique, sans voir le rapport.

Anthropic's Claude Mythos isn't a sentient super-hacker, it's a sales pitch — claims of 'thousands' of severe zero-days rely on just 198 manual reviews https://www.tomshardware.com/tech-industry/artificial-intelligence/anthropics-claude-mythos-isnt-a-sentient-super-hacker-its-a-sales-pitch-claims-of-thousands-of-severe-zero-days-rely-on-just-198-manual-reviews

Please save me a bottle of agi pills @swyx
AI Engineer: 💊 Have you got your @aiDotEngineer AGI Pills prescription from DR @swyx yet?

This is accurate.
And unironically why America wins every new technological era.
Christian Keil: http://x.com/i/article/2037658145213054976

good day

Marc Andreessen breaks down the exact personality profile that separates true innovators from everyone else:
According to him, it's a specific combination of five personality traits, each spiked toward the extreme, that almost never shows up in the same person.
He walks through each one using the Big Five framework:
The first is very high openness. Not just in one domain, but across the board.
This is why many of history's great innovators were unusually creative people outside their main field too, because the trait doesn't discriminate.
But openness alone isn't enough.
&#34;If you're just open, you could just be curious and explore, and spend your entire life reading and talking to people and never actually create something.&#34;
The second trait is extreme conscientiousness, meaning the willingness to apply yourself to a single thing over many years.
@pmarca is pointed about how this reality gets buried under myth:
&#34;The stories told about these people... there's this kid, this stroke of genius, this moment in time... And it's no — for most of these people it's years and years of applied effort.&#34;
Here's where the profile gets genuinely rare: openness and conscientiousness are opposing traits.
Open people drift. Conscientious people grind. Being extreme in both is vanishingly rare.
The third requirement is high disagreeableness. The innovator must be able to hold their conviction when everyone around them says the idea is stupid.
And the world will say it's stupid:
&#34;The reaction most people have to new ideas is, 'Oh, that's dumb.'&#34;
An agreeable person folds and stops pulling the thread. The innovator keeps going, not out of arrogance, but because they structurally don't bend to social pressure.
The fourth trait is high IQ, which Andreessen treats as the price of entry:
&#34;It's hard to innovate in any category if you can't synthesize large amounts of information quickly.&#34;
The fifth and final trait is relatively low neuroticism. The innovator has to withstand years of hard work with no guaranteed outcome, and too much anxiety makes that simply unsustainable.
The bottleneck in innovation was never capital, or the quality of ideas — it was always the rarity of a single person carrying all five traits at once.
And on the rare occasion that person exists, that's exactly where breakthroughs come from.

Just to be clear, there is fraud waste and abuse happening in California on an epic scale. And, it would've never been exposed so emphatically if not for &#34;citizen journalism.&#34; Just because conservatives were the one's who exposed the fraud, it doesn't mean it isn't true. 🤷‍♂️
CBS News: The California Department of Justice and several other state agencies arrested five people Wednesday in a crackdown on an alleged hospice fraud ring. State Attorney General Rob Bonta said the alleged fraud had cost California $267 million through bogus charges to Medi-Cal.
CBS

🚨🇫🇷🇪🇺 This is big: the French government and agencies are officially getting out of Windows & non-EU tech.
Each ministry has to present their exit plan before Autumn: collaboration tools, antivirus, AI, databases..
It's starting with the Digital Ministry dropping Windows for Linux across its own infrastructure.
It's the latest step from France to reduce its tech dependency as much as possible. Previously:
> 80,000 French social security agents migrating to sovereign tools: Tchap, Visio, FranceTransfert.
> the national health data platform moving to a European cloud solution by end of 2026.
There will also be a dependency mapping across all public procurement, and a definition of what counts as a &#34;European digital service&#34; (which could be followed by other EU states as well)
This is coming from the Prime Minister's initiative, signed off by three ministers.
They call this move #GAFAMdetox
Direction interministérielle du numérique: L'État accélère son virage vers encore + de souveraineté numérique 🛡️
Retour sur le séminaire interministériel #SouverainetéNumérique qui s'est tenu hier à Paris visant à réduire les dépendances extra-européennes de l'État. #GAFAMDetox
En savoir plus → https://www.numerique.gouv.fr/sinformer/espace-presse/souverainete-numerique-reduction-dependances-extra-europeennes/

说实话，今天看到这个，我直接把手里所有其他AI记忆方案全停了🤩🤩🤩
YC总裁Garry Tan，把自己天天在用的生产级AI Agent记忆系统，完整开源了
这是他自己跑了很久的真实配置，管着10000+Markdown文件，3000+人物档案，13年的日历数据，5800条苹果笔记，还有所有的会议记录、原创想法
现在他把这套东西打包成了GBrain，MIT协议，所有人都可以免费抄作业
github 地址老规矩评论区自取👇
Garry Tan: If you want your OpenClaw or Hermes Agent to be able to have perfect total recall of all 10,000+ markdown files, GBrain is here to help.
It's exactly my OpenClaw/Hermes Agent setup. MIT-licensed open source. Hope it helps you build your mini-AGI.
https://github.com/garrytan/gbrain

i have another announcement to make.

T-7 days 🇫🇷
Open-source AI Art takes over Paris.
3 days. Hackathons. Art. Talks.
120 spots per day.

We have @SamMorrowDrums talking about the challenges of building and scaling the @github MCP server at @aiDotEngineer 👀

Are you AGI pilled? Dr @Swyx is cooking!

I am using fewer and fewer apps, more and more openclaw and occasional CLI. It's all @steipete's fault. Thank you.

MCP apps coming to GitHub MCP Server!
From the @aiDotEngineer session from Sam Morrow

You know you are at the right community when the head mr @swyx is handing out stickers. At @aiDotEngineer London.

C'mon @sfchronicle : the $1.6 billion &#34;reduction&#34; in the projected cost of California High-Speed Rail between 2024 and 2026 comes from down scoping the project: Merced station no longer in Merced proper and substantial single-tracking which means less frequent service.

approved by @GergelyOrosz !
(Gergely successfully prompt injected the phone booth voice agent by asking it to ignore previous instructions lol)
Boris Starkov: i connected elevenlabs voice agent to a retro rotary phone, and put it in a red british telephone box
its currently exhibited at London AI Engineer summit, and quizzes the attendees on the history of British AI!

If Martin is right, he also just wrote the product spec for open source + distributed compute where broad swaths of groups, individuals and organizations contribute their compute resources to training runs for large param open source models.
There are lots of issues in figuring this out: homogeneity vs heterogeneity of the training clusters, orchestration, financial incentives etc etc etc but some early projects are good signal as to where this can go and that these limitations can be overcome (folding@home, Venice, Tao).
An attempted oligopoly on intelligence is the perfect boundary condition for a bottoms up uprising of fully open, fully distributed AI.
martin_casado: It's only a matter of time before only the model creators have access to the most powerful models. The rest get access to smaller, distilled versions. Or access the models through first party apps and services that don't provide direct access to the token path.
The investment

Re @steipete Imagine building a business over Anthropic models.

As I build my own 2nd brain 🧠 on Obsidian using @karpathy ‘s wiki idea, it suddenly dawned on me - one day when we r gone, our kids could inherit an interactive map to your mind, passion, obsessions, work, fascinations…
It’s kind of beautiful way to think abt your 2nd 🧠.

Today we’re announcing a partnership between Replit and Accenture.
Accenture is investing in Replit, adopting it internally, and working with us to bring secure vibecoding to enterprises globally.
They’re one of the largest companies in the world, with 700,000+ employees and clients across every part of the economy.
The way software gets built is changing. Every company will need to reinvent how they build and operate.
This partnership helps accelerate that shift.
The future of work is about breaking down barriers, and turning everyone into builders.

Having some fun continuing to vibe code the MCP App for @aiDotEngineer with my @OpenClaw, McClaw, that I started in my talk. WIP. Don't judge me and McClaw. 😅

This is the truth. Autocompaction in Codex is so good I never think about context windows anymore.
Daniel Steigman: Pro tip: keep the same thread running as long as you want codex will handle context for you. My oldest thread is 60+ days old and over 59MB

Re Yes it's the tractable form of brain upload. There's a ton of scifi on brain uploads that requires way too exotic tech (scanning and simulating brains etc), when we're about to get a lossy and approximate version of that *a lot* sooner via LLM simulators. You can easily imagine a &#34;brain upload&#34; startup - you show up for a few days to carry out detailed video interviews, then they use all that data with an LLM finetuning process to &#34;upload&#34; you and give you an API endpoint of your simulation that you can talk to. Look at what's already possible with HeyGen as an example, but combine it with an LLM model that has deep knowledge and personality. Trippy and admittedly kind of dystopian but in principle quite possible around now.

European leaders are learning there is no point trying to appease Trump. He has an insatiable appetite for flattery no matter how transparently false. Like any bully, he sees weakness as an invitation to attack again. He respects only coordinated strength. https://trib.al/0QqcfWw

Great fireside chat between @GergelyOrosz and @artman at @aiDotEngineer

Dear MAGA,
Just out of curiosity: If president Biden had banged porn stars, cheated on multiple wives, lied 30,000 times, singlehandedly and unilaterally started a war, released 5,000 Talibanis, bombed a school, bungled the Covid response, added trillions to the deficit, claimed windmills cause cancer, abandoned veterans, bombed 8 countries in one year, altered a weather map with a Sharpie, praised Allah in an Easter message, slept through numerous meetings, threatened to bomb another nation back to the Stone Ages because 'they're animals', violated the Constitution’s emoluments clause, sided with Vladimir Putin over U.S. intelligence agencies, revealed highly classified information to Russian officials in the Oval Office, and golfed when dead service members returned home, would you think it was okay and support his right to do so?
If not, then why let Trump get away with it?

&#34;I hate being on my phone. But I also want to get stuff done.&#34;
Every COO @bran_don_gell's breakthrough moment with his personal agent, Zosia.

If I had a pound for every time a MAGA account bellowed these mad MAGA myths me, I could fund European defence myself.
1. &#34;Europe would be conquered in a week without America&#34;
Europe has 1.7 million troops, 515 nuclear warheads, five aircraft carriers and 1,400 combat jets. Russia couldn't take 20% of Ukraine in four years with a million soldiers. But sure, a week.
2. &#34;Europeans live in no-go zones run by Sharia law&#34;
The &#34;no-go zones&#34; are normal neighbourhoods with kebab shops. The most dangerous thing about them is the parking.
3. &#34;Europeans have no freedom of speech&#34;
Europeans can criticise their governments, protest in the streets and mock their leaders on live TV without getting tear-gassed or shot. In America, immigration agents check your social media, ICE arrests people for their opinions and deports them without trial. &#34;Free speech&#34;, American style.
4. &#34;Europeans pay 60% tax and get nothing for it&#34;
Europeans pay tax and get free healthcare, subsidised university, a year of paid parental leave, 5 weeks holiday and a pension they can actually live on. Americans pay tax and get aircraft carriers.
5. &#34;Europe has open borders and anyone can walk in&#34;
The EU's border agency Frontex has 10,000 guards and some of the strictest visa requirements on earth. The difference is Europe doesn't cage children on camera to win votes
6. &#34;Europe would be speaking German without America&#34;
The British Empire fought Germany for over two years before America joined. The Soviet Union lost 8.7 million soldiers fighting Hitler. The French Resistance ran the intelligence network that made D-Day possible. America showed up in Dec 1941. The war started in 1939.
Every single one of these is repeated daily by people who've never owned a passport and likely never will.

This is anti-competitive.
Peter Steinberger 🦞: Yeah folks, it's gonna be harder in the future to ensure OpenClaw still works with Anthropic models.

Re @swyx 👀 new Chief AI Officer · UK Government 👀
swyx 🇬🇧: gave myself the last keynote of the day.
tune in

I do Sunshine requests so you don't have to!
Jackie Fielder's office claims they are not the source of the leak of a confidential memo.
But texts I obtained have Fielder's aide texting with the reporter who wrote the story, about the subject of the memo, on the day of the leak.

How is LAUSD spending $47K per student?
Enrollment is down 29%, while non-teaching staff is up 42%.
As @skepticaliblog notes, LAUSD has been on a hiring spree that is driving the school district bankrupt.
Kenneth Schrupp: Los Angeles Unified School District spends $47K per student, while just 27% its students meet California’s basic science standards.
Never let them tell you they need more money.

The vibes at @aiDotEngineer in London were really good. The nice (and from what I remember untypical) London weather played very well along. Thanks to @swyx for making it happen and everybody who organized this.

Proselytizers of an apocalypse cult bear moral and legal responsibility for violence committed by their followers.

And that's a wrap! AI Engineer Europe 2026 has concluded.
Our video crew did incredible work to capture the energy, enthusiasm, and positivity of this event -- but it still doesn't come close to being there.
If you're engineering the future of AI -- we hope to see you at a future event!
Thank you London -- thank you Europe!

Claude for Word is now in beta.
Draft, edit, and revise documents directly from the sidebar. Claude preserves your formatting, and edits appear as tracked changes.
Available on Team and Enterprise plans.

Claude for Word is now in beta.
Draft, edit, and revise documents directly from the sidebar. Claude preserves your formatting, and edits appear as tracked changes.
Available on Team and Enterprise plans.

Lyria by @GoogleDeepMind is next level
it did average Turkish pop song very well all my Turkish friends in the conference hall was summoned lol

he's right

Had such an amazing time speaking and meeting amazing people at the conference.
Kudos to @swyx for doing this
AI Engineer: And that's a wrap! AI Engineer Europe 2026 has concluded.
Our video crew did incredible work to capture the energy, enthusiasm, and positivity of this event -- but it still doesn't come close to being there.
If you're engineering the future of AI -- we hope to see you at a

Green and red have never looked so good. 💚🦞
NVIDIA CEO Jensen Huang surprises our panel guests at the #NVIDIAGTC keynote pregame.
Watch the replay to hear the full conversation with @steipete, creator of @OpenClaw, on their partnership and the future of intelligent AI model.
🔗 https://www.youtube.com/live/RTmSrIFZanc?t=6792s

People of pi. You can view my 18 minutes @aiDotEngineer Europe talk here.
May it spark joy. And laughter.
https://www.youtube.com/live/_zdroS0Hc74?t=3632&si=8ovBxJ41FnGMtPO8

People of pi. You can view my 18 minutes @aiDotEngineer Europe talk here.
May it spark joy. And laughter.
https://www.youtube.com/live/_zdroS0Hc74?t=3632&si=8ovBxJ41FnGMtPO8

In the middle of a historic mission back to the Moon, this Administration is proposing to a 47% cut to NASA science and a 23% cut to NASA’s budget overall. Last week’s launch showed our country and world what we’re capable of when we work together toward a common goal. None of that can happen without support for science and research.

Replit now deploys directly to Databricks.
Your apps run inside your Databricks environment while inheriting its security, governance, and data access.
Beta is live. Enterprises are already building with it and seeing massive acceleration in BI and internal tools.

MCP is very much alive! With co-creator of MCP @dsp_ at AIE London

Making moves!
Dan Shipper 📧: first @every app using Claude Managed agents!!
@TrySpiral

today, a tall guy in a colorful sweater walked up to me. i was already at the end of my social energy reserves, having dozens of people walk up to me, never having 5 minutes to breath.
the guy just wanted to say hi and thank me for pi. i thanked him for his kind words, like i did a lot of times today, hoping that i finally get my 5 minutes. until i looked at his name tag.
it was @lucasmeijer one of the most instrumental people behind the Unity game engine. i was sort of star struck and we ended up speaking for 4 hours, chilling outside the venue, finding out that while we never meet IRL we had an immense amount of shared history.
we reminisced about AOT compilers we worked on, shared ex-business partners , our shared emotional rollercoasters when we had to let go of the technical achievements that defined our lives, burning out, finding your identity again, all paired with an immense amount of laughter and the joy of having found a kindred spirit.
we eventually went on a hilariously inefficient pub hunt, ending up at the weirdest fucking &#34;upper class&#34; establishment called &#34;The OWO&#34; where everyone was looking at us two dorks like we were aliens. They made us put stickers over our smartphone cameras and sold us lagers for £11 a pop. hilarious.
i now remember why i loved speaking at conferences 15 years ago. i think i made a friend today.
so, thanks @swyx and crew for putting together @aiDotEngineer and letting me clown around on stage. fantastic vibes, great people.

Trump's pointless war-of-choice just cost an estimated $1 trillion -- enough to provide universal pre-K in America for all 3- and 4-year-olds, or make college accessible to every family earning $125,000/year or less, or reverse the recent health care cuts. https://trib.al/OFI8CIx

It’s a wrap @aiDotEngineer Europe!
Thank you @swyx and team for gathering the best people in the industry in London 🇬🇧 and for such attention to detail and care for the speakers and attendees 👏
Already Looking forward to next year’s!

🫡 @aiDotEngineer in London was extremely good. Absolutely incredible speaker line up, great curation, escaped the Silicon Valley bubble masturbation. Would go again.

To the survivor who risked everything to come forward – I believe you. 
To the Democratic Party – you’d better hold him accountable. If we don’t, we have no credibility asking anyone else to do the same. 
To @ericswalwell – drop out. https://www.sfchronicle.com/politics/article/eric-swalwell-allegations-22198271.php

NEW: Multiple people have resigned from @ericswalwell's campaign, amid a new report alleging sexual assault https://www.politico.com/news/2026/04/10/swalwell-campaign-imploding-amid-sexual-assault-allegation-00867619

Big ups to @swyx and crew. What an amazing @aiDotEngineer Europe! Thanks so much for the opportunity to speak and it was so great hanging with new and old friends!

Spiral shipped a new Managed Agent to power our CLI in an afternoon. Here's what we built and how @AnthropicAI's Managed Agents enabled it. 🧵

what a week! @swyx and team built the room, the builders filled it. never been a better time to be a builder in Europe
AI Engineer: And that's a wrap! AI Engineer Europe 2026 has concluded.
Our video crew did incredible work to capture the energy, enthusiasm, and positivity of this event -- but it still doesn't come close to being there.
If you're engineering the future of AI -- we hope to see you at a

This time around at @aiDotEngineer I was lucky to do something I don’t get to do very often: sharing a keynote! I love that @cristinaponcela introduced herself as “AI native engineer”as someone who entered the industry post ChatGPT. Grow the next generation of programmers people!

This time around at @aiDotEngineer I was lucky to do something I don’t get to do very often: sharing a keynote! I love that @cristinaponcela introduced herself as “AI native engineer”as someone who entered the industry post ChatGPT. Grow the next generation of programmers people!

I strongly suspect that Claude Mythos is a looped language model, as described in the paper &#34;Scaling Latent Reasoning via Looped Language Models&#34; from ByteDance
The authors of that paper called out graph search as one of the areas where looping provides a huge theoretical advantage over standard RLVR. And look at where Mythos blows out its competitors the most

I strongly suspect that Claude Mythos is a looped language model, as described in the paper &#34;Scaling Latent Reasoning via Looped Language Models&#34; from ByteDance
The authors of that paper called out graph search as one of the areas where looping provides a huge theoretical advantage over standard RLVR. And look at where Mythos blows out its competitors the most

The most awesome AI conference @aiDotEngineer by @swyx - doing one of the keynotes for it was the coolest thing! Loved that they brought it to London 🇬🇧

Los Angeles, CA - Rep. Jimmy Gomez statement following the release of a San Francisco Chronicle story:

I've never seen a team ship like @rork
I used to work at @unity where we helped democratize mobile game creation enabling any developer to create a game... But you had to know C# and it would take weeks
Rork literally fulfils the mission. Enabling anyone to build a mobile app in minutes
Congrats to @daniel_dhawan @levan and team!
Rork: Big news: Rork raised a $15M seed round led by @leftlanecap with a16z @speedrun, @peakxvpartners, @trueventures, @GoodwaterCap, and others.
Our team has been building mobile apps for the App Store since we were kids. Now we're making it possible for everyone else to do the same.

I've never seen a team ship like @rork
I used to work at @unity where we helped democratize mobile game creation enabling any developer to create a game... But you had to know C# and it would take weeks
Rork literally fulfils the mission. Enabling anyone to build a mobile app in minutes
Congrats to @daniel_dhawan @levan and team!
Rork: Big news: Rork raised a $15M seed round led by @leftlanecap with a16z @speedrun, @peakxvpartners, @trueventures, @GoodwaterCap, and others.
Our team has been building mobile apps for the App Store since we were kids. Now we're making it possible for everyone else to do the same.

And that’s a wrap. What an amazing few days spent with the AI builders of Europe. Thanks to @swyx for bringing us all together.
The journey has only just begun.

The @aiDotEngineer after-after party.
@Prince_Canuma @adrgrondin @krismatterz (so far)

I have read the San Francisco Chronicle’s account and I am deeply distressed by its allegations. This woman was brave to come forward, and we should take her story seriously.
I am withdrawing my endorsement immediately, and believe that he should withdraw from the race.
San Francisco Chronicle: A woman who worked for nearly two years for Rep. Eric Swalwell, a leading candidate for California governor, said she had sexual encounters with him while he was her boss and alleged he twice sexually assaulted her when she was too intoxicated to consent. https://www.sfchronicle.com/politics/article/eric-swalwell-allegations-22198271.php?taid=69d95b2d5fcb6f0001a6f9f5&utm_campaign=trueanthem%2B3988&utm_medium=social&utm_source=twitter

Huge news: Rork raised $15M seed led by @leftlanecap with @peakxv, a16z @speedrun, @trueventures, @GoodwaterCap, @karman, @sasmith4, Mento & others
&#34;Let's Rork & Roll&#34; pitch from Oct '25
AI will not replace us. It will give our kids the power to make it in life.

Huge news: Rork raised $15M seed led by @leftlanecap with @peakxv, a16z @speedrun, @trueventures, @GoodwaterCap, @karman, @sasmith4, Mento & others
&#34;Let's Rork & Roll&#34; pitch from Oct '25
AI will not replace us. It will give our kids the power to make it in life.

Codex CLI 0.119.0 is out and I'm excited about two features I worked on:
📋 Ctrl+O now copies the last agent response as markdown - works everywhere, even over SSH. I've been using it nonstop to grab plans for review.
🐛 Zellij support is fixed — no more scrollback bug.
👇

I wrote this early this morning and I wasn't sure if I would actually publish it, but here it is:
https://blog.samaltman.com/2279512

nice AIE weekend bonus:
@barry_zyj and @MaheshMurag have officially become the second AIE Millionaire speakers!!
they reached this milestone 2x as fast as @sgrove, our first AIE Millionaire...
... maybe because they had 2 speakers? The Man-Month may not be that mythical?
AI Engineer: Next up! Claude skills are under-appreciated! @barry_zyj and @MaheshMurag from @AnthropicAI are back at AIE stage this year, to tell us about telling Claude about new skills via simple markdown and code files!

JUST IN: The 'Endorsements' section on Eric Swalwell's campaign site has been removed and returns a 404 error.

Just set up @garrytan's GBrain on our company's entire knowledge base — 8,400+ pages of product docs, eng specs, investor updates, and founder notes. All chunked, embedded, and indexed in Postgres.
The hybrid search (vector + keyword + RRF fusion) is surprisingly good — it finds precisely what I need across thousands of docs.
Built a quick web dashboard on top of my GBrain to show the pipeline in action.
Every founder needs this.
Garry Tan: If you want your OpenClaw or Hermes Agent to be able to have perfect total recall of all 10,000+ markdown files, GBrain is here to help.
It's exactly my OpenClaw/Hermes Agent setup. MIT-licensed open source. Hope it helps you build your mini-AGI.
https://github.com/garrytan/gbrain

Wrapped up the 3-day AI Engineer conference in London. Key takeaways:
Planning & verification > implementation. Humans define what to build and validate it works. AI handles the coding.
The harness matters. Terminal bench showed wrapping/orchestrating an LLM changes output quality dramatically. (claude code was one of the worst performing harnesses!)
Parallel multi-agent systems aren't common yet. Factory (Droid) runs parallel agents in read mode only — writes are serialized for testability. (they should try mcp_agent_mail (https://github.com/Dicklesworthstone/mcp_agent_mail) cc @doodlestein )
Bottom line: everyone's still figuring this out. No one's cracked consistently high-quality AI-assisted software yet — and that's because these tools are evolving incredibly fast.
A big thank you to @swyx and the rest of the @aiDotEngineer team for organizing this event!

We're preparing an Intel release for the Codex App and I'm looking for people on Intel Mac willing to test the app before we make it ™️ official ™️
I'll DM you the link!

“I play with it. I see how it feels. I get new ideas. My prompts change. So to me it's a very iterative approach”
this is my favourite @steipete idea
somewhere at intersection of multi-agent token throughput <> interacting with the real world, and real humans (which is messy af), is the future of software
conversation with @swyx from @aiDotEngineer eu, the 🦞 track was great, thanks guys
https://www.youtube.com/live/O_IMsEg91g8?si=-qdkydiMOInc1-au

sam is a wonderful human. many have worked very hard to make you think otherwise
Sam Altman: I wrote this early this morning and I wasn't sure if I would actually publish it, but here it is:
https://blog.samaltman.com/2279512

The @aiDotEngineer after-after-after party
@WolframRvnwlf @altryne @hellheff @liamcbride @raoufcode & @Prince_Canuma