Natalia Murawski

“I would rather have questions that can't be answered than answers that can't be questioned” - Richard Feynman

[Natalia Murawski @n_murashova]

“Just take a step upward one by one without looking at the top of stairs, and you’ll arrive there before you know it.” - words by https://t.co/y8WVuu5DSB - 58

[Леонид Коган 🇮🇱 @abuviktor1]

ТЕКСТ НЕ МОЙ, С АВТОРОМ НЕ ЗНАКОМ, ПРИСЛАЛИ В ЛИЧКУ.
СТРАННАЯ МУЗЫКА, КАК ПО МНЕ, НО ПОЧЕМУ ТО "ПРИЛИПЛА".
ЧТО СКАЖЕТЕ?
"Иногда самое сильное, что ты можешь сделать, это отказаться менять чернила, чтобы подходить к чужой странице. Эти стихи о том, как найти благодать в странном. Оставайся жидким в мире, сделанном из камня. Это называется "Тихий бунт".
Визуализация была сделана с помощью Midjourney и анимирована с помощью VE. Слова написанные мной. Му��ыка, созданная под моим творческим руководством при содействии Суно."
Келли Бош

[J.K. Rowling @jk_rowling]

If you believe free speech is for you but not your political opponents, you're illiberal.

If no contrary evidence could change your beliefs, you're a fundamentalist.

If you believe the state should punish those with contrary views, you're a totalitarian.

If you believe political opponents should be punished with violence or death, you're a terrorist.

[Igor Babuschkin @ibab]

Today was my last day at xAI, the company that I helped start with Elon Musk in 2023. I still remember the day I first met Elon, we talked for hours about AI and what the future might hold. We both felt that a new AI company with a different kind of mission was needed.

Building AI that advances humanity has been my lifelong dream. My parents left the Russian Federation after the collapse of the USSR in search of a better life for their kids. Life wasn’t always easy as immigrants. Despite the hardships, my parents believed that human values were priceless: values like courage, compassion, curiosity for understanding the world. As a child, I admired scientists like Richard Feynman and Max Planck, who relentlessly pushed the frontiers of physics in order to understand the universe. As a particle physics PhD student at CERN I was excited to contribute to that mission. But the search for new physics was getting harder and harder, requiring bigger and bigger colliders, while new discoveries kept getting fewer. So I began to wonder if superintelligence, not larger colliders, could be the key to unlocking the mysteries of the universe. Could AI develop a consistent theory of quantum gravity? Could AI prove the Riemann hypothesis? In early 2023 I became convinced that we were getting close to a recipe for superintelligence. I saw the writing on the wall: very soon AI could reason beyond the level of humans. How could we ensure that this technology is used for good? Elon had warned of the dangers of powerful AI for years. Elon and I realized that we had a shared vision of AI used to benefit humanity, thus we recruited more like minded engineers and set off to build xAI.

The early days of xAI were not easy. Naysayers told us that we arrived too late to the game, so starting a top AI company from scratch would be impossible. But we believed we could do the impossible. Starting a company from zero required lots of hands-on work. In the beginning I built many of the foundational tools used at the company to launch and manage training jobs. I later oversaw much of the engineering at the company, including Infrastructure, Product and Applied AI projects. xAI’s people are deeply dedicated. Through blood sweat and tears, our team’s blistering velocity built the Memphis supercluster, and shipped frontier models faster than any company in history. I learned 2 priceless lessons from Elon: #1 be fearless in rolling up your sleeves to personally dig into technical problems, #2 have a maniacal sense of urgency.

xAI executes at ludicrous speed. Industry veterans told us that building the Memphis supercluster in 120 days would be impossible. But we believed we could do the impossible. Our goal was to get our training setup running at scale on the Memphis cluster ASAP. Towards the end of our 120 day deadline, we were riddled with mysterious issues with communicating over RDMA between the machines. Elon decided to fly to the datacenter, and we followed. Our infra team landed in Memphis in the middle of the night and got straight to work. After pouring through tens of thousands of lines of lspci output we finally identified a wrong BIOS setting, the root of the problem. Elon was there with us until late into the night. When the training run finally worked, Elon posted our triumph at “4:20am” causing us to laugh out loud. I will never forget the rush of adrenaline that night, and the emotional bonds that we were all in this together. We went to bed feeling like we were living through the most exhilarating time of our lives.

I have enormous love for the whole family at xAI. Our team is truly special - you’re the most dedicated people I’ve ever worked with. Catching up to the frontier this quickly hasn’t been easy. It was made possible by everyone’s diehard grit and team spirit. Thank you to every single person who joined me on this adventure. I want to honor your contributions, your time, your sacrifices, which are never easy. I will always remember working together far into the nights and burning the midnight oil. I will never forget the sacrifices and contributions you’ve made. As I drive away today, I feel like a proud parent, driving away after sending their kid away to college. My heart is brimming with tears of joy, rooting for the company as it grows and matures.

As I'm heading towards my next chapter, I’m inspired by how my parents immigrated to seek a better world for their children. Recently I had dinner with Max Tegmark, founder of the Future of Life Institute. He showed me a photo of his young sons, and asked me “how can we build AI safely to ensure that our children can flourish?” I was deeply moved by his question. Earlier in my career, I was a technical lead for DeepMind's Alphastar StarCraft agent, and I got to see how powerful reinforcement learning is when scaled up. As frontier models become more agentic over longer horizons and a wider range of tasks, they will take on more and more powerful capabilities, which will make it critical to study and advance AI safety. I want to continue on my mission to bring about AI that’s safe and beneficial to humanity. I’m announcing the launch of Babuschkin Ventures, which supports AI safety research and backs startups in AI and agentic systems that advance humanity and unlock the mysteries of our universe. Please reach out at [email protected] if you want to chat. The singularity is near, but humanity’s future is bright!

[Natalia Murawski @n_murashova]

Give this playlist a listen: MARIA STUARDA de Gaetano Donizetti https://t.co/qE7uXOiwaX

[boggster @boggster43]

@Travis_4_Trump I believe this is all planned.

[Cassie Kozyrkov @decisionleader]

Let's talk about AI hype. I'll be honest with you, I've been a little bit embarrassed to be associated with AI recently. Human achievement is incremental, but where's our patience for it?

We got a massive speedup on progress, yet the expectations seem to be that we should be solving all of [fill in your favorite massive challenge for humankind] in a week. No, a day. Yesterday!

Can we calm down, please? AI is just another approach to automation. Automation is not new and it's not "solved."

There is so much good stuff in the AI space, but reality doesn't read like a cheap sci-fi paperback. Whatever you're seeing in the tech news either usually took years or is smoke and mirrors.

My team and I spend a lot of time thinking about how long things would take in a perfect world where we had a perfect technology in our hands. (No one has that. And if I ever say something else in a pitch, it's time to put me out of my misery.)

There's always plenty to solve that's not about the tech. The most important part isn't what we *can* do. The most important part is what we *should* do and why we should do it and for whom...

"You've got to start with the customer experience and work backwards to the technology.... It has to start with 'What incredible benefits can we give to the customer?' Not 'let's sit down with the engineers and figure out what awesome technology we have and then figure out how we can market that."

Steve Jobs said this at WWDC '97, but 26 years later the sentiment holds!

Sure, I'm loving this latest AI wave. It's fun to build things that were previously impossible to build, and it's exciting to theorize on where this will be 15-20 years from now!

But as a consumer — and new CEO — of AI products, I understand the AI disillusionment that's been creeping up recently. People have been promised magic, but LLMs are still just a tool! They can drive incredible results *and* they are imperfect tools that need a lot more work.

Remember your audience and what problem you're trying to solve. Selling to enterprise? Consider their scalability and security needs. Selling to developers? Consider their savvy and how they'll use your product.

Consumers are smart. They want things that actually solve problems. Technology is cool, but the industry wins when products are built to solve specific problems for specific people.

************

Speaking of balanced takes, I'm always delighted to promote webinars by Akmal Chaudhri, Esq. of SingleStore (let's continue to encourage him to stay one of the good guys). Here's his next one:

🚀 Info 🚀

Webinar: Apache Spark + OpenAI for Personalized Banking Services
When: Tuesday, April 16 2024 at 1:00-2:00 PM Eastern Time
Instructor: Akmal Chaudhri of SingleStore
Price: Free

🚀 Sign-up link 🚀

https://t.co/i0MtZH6ktI

As always, registering gives you the option of downloading a recording later if watching it in real time isn't an option.

🍿 What it promises to cover: 🍿

* Learn how to leverage Apache Spark and OpenAI for real-time data processing and analysis
* Discover the latest trends and best practices in AI and big data development for personalized banking services
* Learn how to implement AI models using popular frameworks and libraries
* Get insights into new tools for AI development in the banking industry

#DataEngineering #MachineLearning #AI #Sponsored #TechEd #Webinar #Coding #Ap

[Natalia Murawski @n_murashova]

You've Got to List to Last [Fitzpatrick, Joseph R.]

[Luis Wenus @luiswenus]

We built an AI-steered homing/killer drone in just a few hours

I thought it would be fun to build a drone that chases you around as a game. It just uses an AI object detection model to find people in the frame, and then the drone is programmed to fly towards this at full speed as soon as it detects someone.

This literally took just a few hours to build, and made me realize how scary it is. You could easily strap a small amount of explosives on these and let 100's of them fly around. We check for bombs and guns but THERE ARE NO ANTI-DRONE SYSTEMS FOR BIG EVENTS & PUBLIC SPACES YET.

I was also able to add face recognition to it, and only make it attack someone it knew who was, it could easily identify the person from 10 meters distance.

My bet is that we will see some sort of terror attack using this type of tech within the next few years. You still need some technical knowledge to build this now, but it becomes easier and easier.

We need to build anti-drone systems for civilian spaces asap.

I'm usually an open-source absolutist, but I will not post for now, it's honestly super easy to code but no point enabling.

Disclaimer: this was built on a tiny drone for a game and would not hurt anyone, just wanted to spread awareness around this .

[Дождь @tvrain]

В интернете появился сайт в память об Алексее Навальном. На нем можно зажечь свечу в память об убитом политике.

Мемориальный сайт создали независимые журналисты — для тех. кто не может быть сегодня в Марьино. В акции зажжения свечей уже участвуют больше 350 тысяч человек.

https://t.co/W3k6Pbf9g0

[Natalia Murawski @n_murashova]

there is also no power without responsibility

[Natalia Murawski @n_murashova]

“There is no meaningful responsibility without power. It may be only power over what you do yourself—but increased knowledge, increased wealth, leisure are all increasing the domain in which responsibility is conceivable.” - J.R. #Oppenheimer

[Natalia Murawski @n_murashova]

“There is no meaningful responsibility without power. It may be only power over what you do yourself—but increased knowledge, increased wealth, leisure are all increasing the domain in which responsibility is conceivable.” - J.R. #Oppenheimer

[Wall Street Mav @WallStreetMav]

This is true today more than ever.

Never forget: 🚨🚨🚨

[elvis @omarsar0]

I'm convinced that one of the most effective ways to use ChatGPT is for coding.

It has been so useful personally that I have built a personalized code assistant with custom GPT to use @dair_ai.

As an example, I have designed custom instructions for one of our custom GPTs to accept a zip folder containing the project and files and a short explanation of what the project is about (this can probably be automated too).

Then the code assistant automatically analyses the project and code to have a good understanding of the project.

The code assistant then simply prompts the user what issues to fix, debug, or new features to add. I have customized it to ask certain types of questions and have a specific tone.

And with just two user interactions (uploading the project and giving it context on what to do), I can ask it to do something with the project and get coding.

I can even set the code assistant to different modes like learning/explaining, debugging, or analytical mode. This helps the assistant throughout the session to know how to cater to different situations.

The customization can continue further but just this simple custom GPT that I built in like 15 minutes already saves me a ton of time and I can reuse it to run different sessions.

I work on a lot of open-source projects, so I have been thinking of publicly sharing a more customized version of our code assistant so that our community members have access to it and allow them to learn the code base faster and more easily contribute to our projects.

I share this here not because I think the custom code assistant is novel. I share this because I am absolutely astonished at how powerful these systems have become and how deeply integrated they have become in my workflow, not to mention the potential to elevate our community.

I pay $20 per month for this but I think the value I get from ChatGPT is worth a whole lot more. I use this every day and it allows me to do other creative and deep work which I don't think was possible a year or two ago.

As I share this here, I am already working on a research assistant to help me automate many tasks on which I spend the majority of my time.

Don't get me wrong, these systems are not perfect. But I never use them with that expectation. I actually prefer to customize them to make the best use of them for the actual tasks I need help with. It often turns out that customizing them makes them more useful and accurate. That's why I am a huge believer in the power of personalized assistants.

Would love to hear how others are finding value with custom GPTs?

[Nick Davidov @Nick_Davidov]

Note to self: many Harvard students don’t look at what they are signing and supporting, or if supporting don’t understand exactly what they are supporting. Your future lawyers and politicians, everybody.

[Natalia Murawski @n_murashova]

The Google guys pitched their idea in one sentence of just 12 words. They told Moritz that Google's mission is "to organize the world's information and make it universally accessible and useful."

[Bojan Tunguz @tunguz]

I have so many questions.

[Natalia Murawski @n_murashova]

very interesting art

[Milana Blackman @milana_blackman]

Link in bio for limited edition prints ⬆️