We’re moving Gemini 2.5 Pro to Preview today, giving developers access to increased rate limits to begin testing the model for production-ready apps. It’s available now in Google AI Studio. Learn more ↓ https://t.co/udg5qGpg83
🚨🇺🇸EXPOSED: ANTI-ELON PROTESTS ARE STAGED & PAID—BUSSED-IN, SCRIPTED, CLOCKED-OUT
The anti-Elon, anti-DOGE, anti-Trump protests in D.C.? They aren’t grassroots. They are payroll-driven theater.
- Buses rolled in packed with hired protesters.
- Pre-made signs handed out assembly-line style.
- Scripts distributed to keep messaging “on brand.”
- Protesters all left at once—just like a shift change.
The protests are organized astroturf—NGO-backed, donor-funded, and as fake as their outrage.
It's a union of grifters and bureaucrats trying to stop Elon from cutting off their taxpayer-funded gravy train.
Source: @WallStreetApes
America’s leading AI companies are all reporting that demand is off the charts — so much so that they are being forced to impose rate limits. Fortunately a massive new infrastructure build-out is already underway thanks to President Trump. He is unleashing American energy, making permitting easier, and attracting huge new investment projects. The Golden Age of American AI is just getting started!
On today's Wordle, the new Gemini model completely crushed the competition. It logicially deducted diverse words, found the correct spots of valid and invalid letters and got a result quickly. Sonnet proposed multiple invalid words in the end, so DNF
Gemini 2.5 Pro physics simulations in Three.js!
All of these started out as "one-shot prompts" but I continued to query Gemini for better results.
Clone with GitHub below 👇
#threejs#Physics
Gemini 2.5 Pro is now *easily* the best model for code.
- it’s extremely powerful
- the 1M token context is legit
- doesn’t just agree with you 24/7
- shows flashes of genuine insight/brilliance
- consistently 1-shots entire tickets
Google delivered a real winner here.
For the record I do not bet on this multiyear research fad.
To my understanding, the main way to manipulate the inner workings of AI is representation control. It's been useful for jailbreaking robustness, finetuning resistant unlearning, utility control, model honesty, etc. 🧵