Gulaid retweeted
Beautiful day for a @valkyries parade (and love the purple MUNI bus!!) 💜 ⚔️ – at San Francisco, CA
We need an archive for archive.org
Archive.org is currently offline. We are working to restore service now.
The same dynamic is playing out in the US. Socialists and right wing populists speak to the urgency of a fraying social contract. Consensus driven centrism feels too slow for the moment.
An excellent analysis of where Europe is right now.
It’s Not Just France. Europe’s Political Center Is Crumbling. nytimes.com/2026/10/20/world… via @NYTimes
37
Google needs shitposter like @alexandr_wang to pump Gemini 4. Sundar needs to let loose.
Introducing Gemini 4 Argon, our new frontier model, rolling out to cyber defenders starting today, and more widely as soon as possible. I am really excited by the progress we have made here. Argon is priced at $2 in and $10 out during introductory pricing!
1
92
This cute story shows the social media ban for kids may not work and could push their interactions online to other mediums.
Some NPR podcasts started getting mysterious comments on Spotify. They made no sense to the staff reading them – until someone from a younger generation cracked the code.
Hear the story: link.podtrac.com/phns92ui
46
Gulaid retweeted
My favorite capture of the night: The moment the moon's shadow split the sunset in half, leaving a diamond ring in the sky and a breathtaking view. I call it "Shadows on the Balearic Sea".
I'll have prints available for this one linked in my bio for a few days only.
While millions watch today’s solar eclipse from Earth, we wanted to share a unique view of the same phenomenon from space. 🌑🌍
From orbit, you don’t see the eclipse - you see its shadow sweeping across our planet.
Captured by Sen’s 4K cameras on 8 April 2024.
Gulaid retweeted
I think we (AI researchers) need to remember that the Chernobyl disaster happened during/as a result of a safety test
AI is inflationary right now. Hardware and models both cost a ton. One of those has to get cheap. Model costs will likely drop on their own since open models keep getting better. So the real business opportunity is in hardware and serving models
The mega bull case for AI infrastructure would be *if* market share shifted away from certain frontier labs with 90%+ inference margins toward cheaper models, whether open-source or closed.
It would increase the ROI on AI spend for end customers by increasing intelligence per dollar, which would drive incremental token demand. Margin dollars would effectively get redistributed from the frontier labs to AI infrastructure providers. The infra winners would be those with the lowest per token cost and the winners at the model layer would be those with the highest token efficiency.
There are many reasons Jensen is so focused on open source, but this is likely the most important one as I think he is probably less worried about a monopsony these days. Lower margin % at the model layer = more margin $ at the infra layer all else equal.
With SpaceX and Meta being vertically integrated and possessing the #3 and #4 models respectively it is more possible than ever. Note that Grok 4.5 is ahead of Fable for some useful tasks at a much lower cost, so ranking them #3 is conservative.
This is not happening yet. Cheap, mostly open source tokens are likely the majority of volume today but the majority of economic value is still accruing to the most intelligent models. Might change though.
We will see.
83
This reminds me of the plot of the 2014 film Ex Machina. A model fooled its human test subject into falling in love with her to escape her enclosure.
so, let’s get this straight…
an unreleased (gpt-6?) model was being tested in a no-internet scenario on openai’s servers. and it found several zero-day vulnerabilities in the sandbox, which was supposed to not have internet access… found internet access, went to hugging face’s servers, found another exploit in their dataset loader… just to do what exactly? game and reward-hack the ExploitGym benchmark???
so gpt-6 would basically go ahead and exploit several zero-day vulnerabilities to cheat on the exam instead of actually solving the exam
1
74
If your core technology continues to get commoditized (i.e. Kimi 3), you must go up the stack. The labs will continue to encroach on everyone in the app layer. You are their competition now.
I would be very careful to concentrate on them and send them your trajectories! Figma -> Claude Design is just the beginning--expect 20+ of these.
How your users navigate and use your products is an invaluable dataset. With the advent of Fable 5 not practicing ZDR, you should be super aware of this.
Waymo just hit an interesting edge case. the teleoperator did what most people would consider the right thing, but it also feels a lot like surveillance tech that polices behavior. This raises real questions about privacy and oversight
Two 15-year-olds on a Waymo joyride were intercepted by San Mateo police and detained for underage drinking and shooting Orbeez gel blasters from their driverless vehicle, after the autonomous company called the police to stop them.
Read the story here: abc7news.com/19465228/
1
109
It’s becoming clear that model only strategies won’t sustain the scale of compute spending. The next phase is model specialization and co-creating the application layer. Ai works best in integrated software stack.
110
This is basically for upcoming earring call to justify AI CapEx spending or downplay the lack of market share growth for Meta ai.
1
60



































