2026-06-26

something has definitely shifted in the past few weeks. seeing a huge uptick in large enterprises wanting to secure compute and post-train their own models in house, frequently on top of GLM-5.2. everyone is starting to understand how open source wins.

taking "@openai is cooking" to a new level.
our chief research officer @markchen90 loves to cook so when @swyx and @allenpark started a new show, there was only one thing to do.
https://www.youtube.com/watch?v=fpAthTtha8c&t=5s

Everyday the admin holds back American frontier models is a day the chinese models gain market share and customer trust.
This will be unbelievably damaging to the American economy overall.
zerohedge: "the share of tokens used for US models on OpenRouter has collapsed": Bloomberg

Re Garry, been following you for a bit. I work at Gusto and loved your talk you did with Eddie.
I’ve been using GBrain on my personal OpenClaw for a bit and recently have been adapting it for Gusto’s Claude Code flows (I had built an adapter but just ripped it out for your Claude Code MCP).
Getting GBrain hooked into Gusto’s MCP stack and workflows has veen been a bit of work, but oh my lord it was so worth it.
Claude Code with MCPs to our data sources - nice, but still kind of an idiot.
Claude Code with GBrain indexing my Slacks, meeting notes, documents & dashboards - PRICELESS!!
Now I’m building loops to automate my PM grunt work and focus more time talking to customers. Wouldn’t be doable without your inspiration & GBrain. Thank you 🙏

A funny thing is that LLMs are not particularly good building agentic systems. For the moment, you need human expertise to do it well.

http://x.com/i/article/2069838668370673665

There will be more and more pressure and more money and lobbying behind open soon.
It is not a "two big companies have all the power in AI" situation like some imagine.
The financial incentives are shifting away from single monopoly ownership and those companies will have a lot of power to push back against regulatory capture and overreach. We need this to happen a lot faster before the barriers are setup and destroy us.
That's because the opposite is also possible if we keep going down the road of closed.
The more our companies and governments gate and close off, the worse we compete in the world.
We are single handedly killing our own economy and competitiveness with stupidity right now.
Slowing rollouts. Half baked orders and laws. Outright stupid laws. Fear Mongering. NIMBYs. Luddites running rampant. Regulatory capture.
The future is ugly if we keep going down this path.
We're speed running dystopia.
But the economic winds are shifting. American business will have a lot of say in what comes next. They need to wake up and see that the regulatory capture and populism is not in their favor and that if we have to beg companies and governments for access and submit to face scans and upload our birth certificate, it will work briefly and then it will really start to hurt us, and then we simply will not be able to compete and we'll incentivize the rest of the world to run fast in the another direction.
Rohan Paul: UBS says 60% of companies now watching AI budgets are moving to cheaper models and open-source Chinese models
The pressure is coming from extreme bills, including users spending up to $35K/month, teams exceeding quotas by 200%, and companies cutting internal AI tools from 5 to

It doesn't take much skill to game this out and see where we're going if we keep this up.
Kneecap our own companies. Gate everything. Let regulatory capture and fear mongering win. Have our people and companies beg for access to models from AI labs and Big Daddy government.
It spells the end of our competitiveness, destroys trust, wrecks our economy, destroys beneficial second order effects and gift wraps a turbo booster for China, their economy, and their tech stack.
I have never seen such short-sightedness and stupidity in the county I love more than anything in my entire fifty years.
And I've never seen more morons pushing for this and cheering it on.
We've got to wake the hell up fast.
Perry E. Metzger: What happens if American AI labs can't compete with Chinese labs? You can regulate US labs all you want, you can ban Americans from using open source software, you can do all sorts of counterproductive things, and all of them will just enhance Chinese competitiveness.
The

Re @ZeffMax Because Dario sees himself as the high priest of the AI era, the only one qualified to stand between the most advanced models and users. But this is all his fantasy, arrogance, greed, and foolishness.

“We're in a weird state of the world where the supposedly totalitarian regime is trying to open up the technology, and the supposedly democratic government system is trying to restrict and control technology. We're in, like, opposite world.” - @pmarca 🎯
Inside AI Policy: White House advisor Andreessen opposes export controls on Anthropic’s Fable http://ow.ly/7K4g106A614

Fable is currently export controlled & rumors are that 5.6 will also be subject to an approval framework. Whatever jiu jitsu the Chinese are using to get us slow down our own frontier models while letting their models run free appears to be working. Who is capturing who? 🧐🇺🇸
Bill Gurley: This is what’s causing Anthropic to aggressively beg for govt protection (see below). Customers are finding cheaper alternatives. Keeping employees requires continuing ultra-rich secondaries ($$$) that are dependent on revenue growth. When you can’t win on the field go to DC.

My friend the brilliant @kevin_volpp dropped dead
He survived (read abt it: https://tinyurl.com/ypcpdzpz)
But every year 300-400,000 in the US don't
This week: our @nature article (+A.Schubert J.Ross @m_sendhil M.Lingman) applies AI to this medical mystery
https://www.nature.com/articles/s41586-026-10674-6

The OpenAI Foundation is joining Intercept as a founding partner through our AI resilience program.
As AI accelerates advances in biology and medicine, we also need to prepare for the risks of misuse, including the potential for engineered outbreaks.
Intercept Health Fund is working to end respiratory pandemics through a pathogen-agnostic approach, investing in broad-spectrum preventatives and non-pharmaceutical interventions that can protect against a wide range of threats and strengthen a new layer of resilience for humanity.
https://blog.interceptfund.com/p/ending-respiratory-infections

Congratulations to @TheVOSF for winning 9 CA Journalism Awards, including 4, 1st Place Awards. Please help us expand our coverage & reach by urging your friends & family to subscribe w/ a monthly $10 tax-deductible donation: http://thevoicesf.org/support-our-publication
And special thanks 🙏to the creator of http://911sf.org - who helped me snag 1st place for this article: https://thevoicesf.org/open-data-shows-sf-police-drug-calls-surge-50-in-two-days/

from Write & Learn (thx @MilksandMatcha @philipkiely @swyx), my most important takeaway, 𝐛𝐮𝐢𝐥𝐝 𝐚 𝐝𝐞𝐧𝐬𝐞 𝐫𝐞𝐰𝐚𝐫𝐝 𝐟𝐨𝐫 𝐲𝐨𝐮𝐫 𝐰𝐫𝐢𝐭𝐢𝐧𝐠.
fun fact. most people know @swyx as the famous podcaster, the AIE godfather. meeting him in person, I saw something different ... (clickbait intended)
lily zhang: http://x.com/i/article/2069838668370673665

we're hosting a mango tasting next week at the AI eng conference in SF! 🥭
we're literally importing 100s of mangos from India for this (I had to ask my dad for an invite to his secret whatsapp group)
if you've never had indian mangos before, you're missing out. easily the best fruits in the world
stop by to try them out!
Extend: Our team is bringing literal crates of mangos to the @aiDotEngineer World Fair during the Welcome Reception!
Served freshly cut and perfectly ripe, come by the AIE Lab booth Monday 4-7:30 to see if it lives up to the hype 🥭
We have it on good authority that they will go fast

Growing up I played a ton of Zynga games like FarmVille and Words with Friends (didn’t even realize Words was a Zynga game until years later)
This was a full-circle moment now building in consumer AI and getting this surprise book in the mail!
Excited to dig into Life at the Speed of Play by Mark Pincus and pairing it with the recent podcast too - thank you @garrytan!!
Takeaways so far:
1. Ship at the speed of play
2. Obsess over what people actually love
3. Turn iteration into a repeatable system

Optimizing for social approval from people who aren't building anything is the very best way to stay mediocre.
https://x.com/garrytan/status/2067101283493040518
Garry Tan: You’ll never achieve anything if you are afraid of being cringe
Then again this is from someone trying to use looksmaxxing to achieve outsize return, good luck with that

Forward Deployed Engineering is the most critical ingredient for Enterprise AI adoption. It’s why the most important companies on the planet are investing so heavily into it.
That’s why we put together the Forward Deployed track at this year’s AI Engineer World’s Fair on Tuesday, June 30.
Learn about the current state of the industry and where its headed from the people and companies leading the charge.
Everything it REALLY takes to deploy agents in production and transition our economy to the new world order
Don’t miss it :) (more details and link in comments) ↓

I was at the first AI Engineer Summit @aiDotEngineer ~500 people, limited admission, felt like a secret.
Next week it takes over Moscone West: thousands of engineers, 400+ sessions. Huge props to @swyx. 👏
And this year my team is on the stage. 🎤
July 2 — Akele Reed, Dave Revere & Doug Keller present: "Evals-Driven Development: Engineering a Mental Health AI Coach Ethically & Safely"
In mental health, vibes can be dangerous. So we built both safe and effective: → Clinical Feedback Loop (therapist judgment → machine-readable evals, in prod) → An Ethics Engine of modular guardrails → Supervisor/Executor/Evaluator agents for clinical adherence
Couldn't be prouder. Room 2005, 1:30pm. Come with hard questions.
🔗 http://ai.engineer/worldsfair/2026

So let me get this straight:
1. We publish frontier models
2. They get distilled into cheap open source models
3. US companies adopt the same open source models because they're good enough and much cheaper
4. We start gating access to frontier models
What next? US companies innovate less? Open source models become more attractive?

The decades-long prison sentences imposed on a group of Texas activists who were protesting against ICE, using the guise of "terrorism," are an outrage clearly designed to suppress such activism. https://trib.al/OzfJuFw

Step 1: Invent a new extra-bad category of crime, like "hate crime" or "terrorism", with extra-harsh punishments.
Step 2: Expand the definition of the new kind of crime to include anything, or anyone, that you want to punish extra harshly.