Hello @kritisanon,
1. Don't play the woman card. You were not criticized for your gender, but for your choice of clothes in an ad. You are bringing gender into this to make it a women's issue and seek wider support from women and male feminists. It's not a gender issue. Even Ranveer Singh was criticized on Twitter for a nude photoshoot. He didn't play the gender card.
2. Personal choice stops where public commercialization begins. What you wear off-camera is entirely your prerogative. You've been in this business for years. Until now, nobody bothered you about your choice of clothes. You know why? Because nobody actually cares what you wear. The outrage here comes from a commercial choice: sexualizing a deeply sacred festival celebrated by millions. Context dictates appropriateness. If a male actor appeared bare-chested while portraying the brother in that exact scene, the public pushback would be identical. The issue is sacredness and context, not gender.
3. Claiming "clothes don't define culture" directly contradicts human history. Attire is a foundational visual language of every global tradition, you cannot wear a Kimono and call it Indian ethnic wear, nor can you strip a festival of its sacred aesthetic without altering its significance. If attire truly has zero bearing on cultural respect, then tomorrow is Eid. Can you wear that exact same outfit you wore in that ad and wish Eid Mubarak? You said clothes don't define the respect you have for culture in your heart. So do it and we'll take our criticism back.
Respecting a tradition means honoring its context, not commodifying it under the guise of fashion.
In my ~14 months at @OpenAI, one of the most surprising and genuinely wonderful things about it, has been the full leadership support for full-duplex models (gpt-live series), even when it seemed deeply improbable that they would work.
And yet, if they did work, it was obvious how magical they could be. I am quite sure this is not an exception but a general rule for research projects at @OpenAI.
There were so many moments when the problem felt impossibly hard. But one thing kept being true: it was never clear why it shouldn’t work. And almost every time we understood the problem a little better, it became a little easier to solve.
I’m pretty sure there are very few environments in the world where a bet like this could have been made and sustained. So grateful to be part of this wonderful place.
@demishassabis We figured out how to make sand think, and our plan is to race each other off a cliff. This is the sanest take I’ve read on AI. We have one shot to get it right.
I built a Claude Code skill that treats your code as a state machine, not a text file.
It auto-activates on concurrent/async code and scans for 12 hazard patterns: → Deadlocks → TOCTOU races → Priority inversion (crashed Mars Pathfinder) → Zombie states + resource leaks → Unguarded shared mutation → ...and more
Each finding has a confidence score. It generates test harnesses that force the dangerous interleaving to happen. And it only raises the alarm when it's worth your attention.
Silence = safe. That's the signal.
Inspired by a weird question: what would Turing want from an AI dev tool? Probably not faster autocomplete.
Installable in Claude Code
https://t.co/6RgJ0BWJPW
@pritish_yuvi@sama@OpenAI OpenAI’s nonprofit can sponsor them too. Not jobs, just emotional support and a yearly email saying “your code made a difference.
@sama Incredible that before 2024, developers had to type entire functions BY HAND. With their FINGERS. No generation has sacrificed more. I’m told some of them even had to read documentation.
Meta quietly delayed their next model after testing it against Gemini 3.0.
Not because it needed polish.
Not because the timing was wrong.
It simply couldn’t beat it.
Avocado was benchmarked against Google on reasoning, coding, and writing. The result was clear enough that Meta pulled the launch.
Think about what that means.
Google is now part of Meta’s internal release checklist.
And that says more about Gemini than it does about Meta.
Interviewer:
X has 200M+ active users posting simultaneously.
If Ronaldo with 80M followers tweets,
how do you update 80M feeds instantly
without melting your cache clusters?
Announcing new open-source Python package: aisuite!
This makes it easy for developers to use large language models from multiple providers. When building applications I found it a hassle to integrate with multiple providers. Aisuite lets you pick a "provider:model" just by changing one string, like openai:gpt-4o, anthropic:claude-3-5-sonnet-20241022, ollama:llama3.1:8b, etc.
pip install aisuite
Open-source code with instructions: https://t.co/gwz9oKTCFx
Thanks to Rohit Prsad, Kevin Solorio, @standsleeping, Jeff Tang and @Johnsanterre for helping build this!
Feel for Rohit Sharma.
Wasn’t selected for 2011. Couldn’t be a World Champion. Publicly expressed disappointment.
12 yrs later, did everything right. Nearly saw us through. But life is tough. Age is catching up. This will be his last one. Well played Captain.
The AC was not working, KK was sweating profusely, then when he had chest pain, this is how he was taken out. No paramedics for such a big concert, no stretcher. They made him walk during a cardiac arrest.
Suddenly ex-Pakistani players are teaching Indian fans how should they treat their players. This holier than thou approach by these bigots who didn’t let Danish Kaneria pick food from the same table because of his faith.
Most hilarious.
To be a part of a mythology as a college student, and as a Father now, believing in the legend, believing that he can still finish matches by hitting sixes & fours, the legend called MSD. We are honoured.