正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Working with Physical Intelligence, we are bringing general purpose robotics models into the places that need them most —- the factories that make things we all need and use. 150 factories using Formic robots and counting!

Saman Farid

4,156 subscribers

18,044 次观看 • 24 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

14 条评论

Saman Farid 的头像

Saman Farid24 天前

we started @goformic with the fundamental conviction that robot intelligence will get better and cheaper rapidly. But deploying at scale required building out physical infrastructure, capabilities and customer trust to deliver high performance. We’ve now earned the trust of customers across hundreds of sites; where we are delivering 99%+ uptime day-in day-out.

Saman Farid 的头像

Saman Farid24 天前

The best robots in the world are not doing backflips - they’re boring and reliable. We’ve seen through real deployment experience that this will need to be a MIX of specialist and general purpose robots. And we’re excited to be able to bring BOTH - to solve customers biggest problems.

Saman Farid 的头像

Saman Farid24 天前

Robotics users are pragmatic — they want whatever works. — not just cool demos.

Quan Vuong 的头像

Quan Vuong24 天前

@physical_int Excited to partner with you @samanfarid !

Ali Tamaseb 的头像

Ali Tamaseb24 天前

@physical_int Awesome to see!

Shamoon Siddiqui 的头像

Shamoon Siddiqui24 天前

@physical_int Congrats!! 150 factories is a ton of real floor time. Are you running the Pi models cloud-hosted or on-site? Pi's writeup flagged connectivity as an issue for some deployments and factory wifi is not always a friend

clankr 的头像

clankr24 天前

@physical_int Congrats. So how the latency is handheld on API?

Chetan 的头像

Chetan23 天前

@physical_int We need. More. Robots. 🔥

👍🏻 的头像

👍🏻24 天前

@physical_int So. Many. Bots.

Saman Farid 的头像

Saman Farid24 天前

@physical_int Here’s to 100,000 more

Lorimer Ventures 的头像

Lorimer Ventures23 天前

@physical_int

Andile (Ethan) 的头像

Andile (Ethan)23 天前

@physical_int Any interest in different grippers if they were faster lead time and cheaper?

Amine Kharrat 的头像

Amine Kharrat24 天前

@Stone_Tao @physical_int Oha Saman, unexpected to see formic following the humanoid route

plastic populism 的头像

plastic populism23 天前

@physical_int manger at Amazon would shoot me on sight if he see me work this slowly . Fuck out of here . Cheapest labor always going to be human labor because you can gonna threaten to replace them with this crap and people have can take that pay or just starve or gamble

相关视频

Huang: We used to call them data centers but the things that we're building now, these are really SI factories, super intelligence factories

Huang: We used to call them data centers but the things that we're building now, these are really SI factories, super intelligence factories

Acyn

716,949 次观看 • 1 个月前

We need to get out of this mindset that wars are won by soldiers alone. They're won by supply chains, factories and the country behind them. That's the conversation we need to have.

We need to get out of this mindset that wars are won by soldiers alone. They're won by supply chains, factories and the country behind them. That's the conversation we need to have.

Al Carns

191,445 次观看 • 4 个月前

One of the biggest surprises in AI over the last few years has been how well coding agents generalize beyond software. In a recent essay, MIT professor Philip Isola argued that we may be entering the era of robot-use agents: general-purpose models that can control different robots, write policies, and learn new physical tasks with little or no robot-specific training. In this episode of Decoded, we're joined by the founders of Waddle Labs and Robocurve PBC, two of the startups whose work helped drive this realization. They're working at the frontier of using general-purpose models to control robots, and their recent demos helped inspire the growing conversation around robot-use agents. Together, we dig into the research behind that idea, from code-as-policies and vision-language-action models to the harnesses and evals needed to make these systems work in the real world. 00:00 — What Are Robot-Use Agents? 02:11 — From Vision-Language-Action Models to General LLMs 04:07 — The Bitter Lesson for Robotics 07:04 — How Coding Agents Learned to Control Robots 10:41 — How Robots Learn From Experience 14:22 — The Harness as a Form of Robot Intelligence 16:14 — Watching Astra Control a Robot 20:55 — Why General Models May Win in Robotics 26:05 — How Close Are General-Purpose Robots?

One of the biggest surprises in AI over the last few years has been how well coding agents generalize beyond software. In a recent essay, MIT professor Philip Isola argued that we may be entering the era of robot-use agents: general-purpose models that can control different robots, write policies, and learn new physical tasks with little or no robot-specific training. In this episode of Decoded, we're joined by the founders of Waddle Labs and Robocurve PBC, two of the startups whose work helped drive this realization. They're working at the frontier of using general-purpose models to control robots, and their recent demos helped inspire the growing conversation around robot-use agents. Together, we dig into the research behind that idea, from code-as-policies and vision-language-action models to the harnesses and evals needed to make these systems work in the real world. 00:00 — What Are Robot-Use Agents? 02:11 — From Vision-Language-Action Models to General LLMs 04:07 — The Bitter Lesson for Robotics 07:04 — How Coding Agents Learned to Control Robots 10:41 — How Robots Learn From Experience 14:22 — The Harness as a Form of Robot Intelligence 16:14 — Watching Astra Control a Robot 20:55 — Why General Models May Win in Robotics 26:05 — How Close Are General-Purpose Robots?

Y Combinator

95,405 次观看 • 1 个月前

We will not use taxpayers' money to fund a cathedral that has no benefit to the ordinary Ghanaian. NEVER! We need schools, hospitals, factories, and Good roads 💪

We will not use taxpayers' money to fund a cathedral that has no benefit to the ordinary Ghanaian. NEVER! We need schools, hospitals, factories, and Good roads 💪

Elorm🤞♌️

64,435 次观看 • 1 个月前

Robots can already fold laundry, make espresso, clean kitchens, and assemble things. The harder problem is getting them to do those tasks reliably, for long periods of time, without a human babysitting them. At Startup School 2026, Physical Intelligence cofounder Chelsea Finn explains what it takes to build general-purpose robots that work in the real world. She shares how reinforcement learning pushed robot throughput up 2x, how their systems can run autonomously for hours, and why she believes robotics is entering its GPT era: moving from specialized models toward general-purpose systems that can work across tasks, robots, and environments. 00:00 — The State of Physical Intelligence 01:23 — What It Takes to Make Robots Useful 05:11 — The Reliability Problem 07:43 — Reinforcement Learning for Robotics 09:35 — Learning From Failures 12:43 — Training Robots to Improve Themselves 14:21 — Can a Robot Work for 13 Hours Straight? 17:36 — Why Robots Need Memory 21:22 — Building a General-Purpose Robot 25:02 — From Fine-Tuning to Out-of-the-Box Models 27:35 — Training on All the Data 30:20 — One Model That Beats the Specialists 31:21 — Compositional Generalization 37:49 — The GPT Era of Robotics 39:49 — Q&A

Robots can already fold laundry, make espresso, clean kitchens, and assemble things. The harder problem is getting them to do those tasks reliably, for long periods of time, without a human babysitting them. At Startup School 2026, Physical Intelligence cofounder Chelsea Finn explains what it takes to build general-purpose robots that work in the real world. She shares how reinforcement learning pushed robot throughput up 2x, how their systems can run autonomously for hours, and why she believes robotics is entering its GPT era: moving from specialized models toward general-purpose systems that can work across tasks, robots, and environments. 00:00 — The State of Physical Intelligence 01:23 — What It Takes to Make Robots Useful 05:11 — The Reliability Problem 07:43 — Reinforcement Learning for Robotics 09:35 — Learning From Failures 12:43 — Training Robots to Improve Themselves 14:21 — Can a Robot Work for 13 Hours Straight? 17:36 — Why Robots Need Memory 21:22 — Building a General-Purpose Robot 25:02 — From Fine-Tuning to Out-of-the-Box Models 27:35 — Training on All the Data 30:20 — One Model That Beats the Specialists 31:21 — Compositional Generalization 37:49 — The GPT Era of Robotics 39:49 — Q&A

Y Combinator

114,240 次观看 • 2 个月前

I visited the world's #1 humanoid robotics company in Shanghai. Most robotics demos are scripted, controlled, and secretly operated by humans. This 3-year-old company shipped 10,000 robots last year. Into real factories.

I visited the world's #1 humanoid robotics company in Shanghai. Most robotics demos are scripted, controlled, and secretly operated by humans. This 3-year-old company shipped 10,000 robots last year. Into real factories.

Shruti

42,566 次观看 • 4 个月前

CORRECTION: The Chinese government is pouring subsidies into developing humanoid robots that are becoming cheap and smart enough to replace human workers in factories. We are deleting a post containing incorrect information

CORRECTION: The Chinese government is pouring subsidies into developing humanoid robots that are becoming cheap and smart enough to replace human workers in factories. We are deleting a post containing incorrect information

Reuters

42,670 次观看 • 1 年前

.Kush Desai: "Yesterday, we saw new orders of core capital goods — which are the industrial machinery, the things you need to make manufacturing facilities and factories — those orders actually exceeded expectations by over sixfold."

.Kush Desai: "Yesterday, we saw new orders of core capital goods — which are the industrial machinery, the things you need to make manufacturing facilities and factories — those orders actually exceeded expectations by over sixfold."

Rapid Response 47

88,182 次观看 • 6 个月前

“In 2024, Chinese factories installed roughly 300,000 new industrial robots. American factories installed 30,000. Today, there are now more than 2 million industrial robots working in Chinese factories. In the U.S., the number is one-fifth that,” says Rush Doshi, director of CFR’s China Strategy Initiative and assistant professor at Georgetown University’s Walsh School of Foreign Service. “China is investing in the future, it’s pulling ahead in research publications and patent filings, and pouring hundreds of billions of dollars into the robotics sector.”

“In 2024, Chinese factories installed roughly 300,000 new industrial robots. American factories installed 30,000. Today, there are now more than 2 million industrial robots working in Chinese factories. In the U.S., the number is one-fifth that,” says Rush Doshi, director of CFR’s China Strategy Initiative and assistant professor at Georgetown University’s Walsh School of Foreign Service. “China is investing in the future, it’s pulling ahead in research publications and patent filings, and pouring hundreds of billions of dollars into the robotics sector.”

Council on Foreign Relations

22,853 次观看 • 6 个月前

Chelsea Finn (Chelsea Finn) on building general-purpose robotics, and bringing intelligence into the physical world. At AI Startup School in San Francisco. 00:00 - General Purpose Robots 00:11 - Challenges in Robotics Applications 00:57 - Physical Intelligence: A New Approach 01:47 - Learning from Language Models 02:08 - Data Sources for Training Robots 03:32 - Training with Real-World Data 04:39 - Initial Successes and Challenges 09:10 - Breakthrough in Robot Training 11:03 - Improving Performance 15:43 - Expanding Capabilities 17:34 - Robots in Unseen Environments 25:54 - Handling Open-Ended Prompts 29:36 - Evaluating Robot Performance 30:03 - Future Directions and Challenges 31:27 - Audience Q&A

Chelsea Finn (Chelsea Finn) on building general-purpose robotics, and bringing intelligence into the physical world. At AI Startup School in San Francisco. 00:00 - General Purpose Robots 00:11 - Challenges in Robotics Applications 00:57 - Physical Intelligence: A New Approach 01:47 - Learning from Language Models 02:08 - Data Sources for Training Robots 03:32 - Training with Real-World Data 04:39 - Initial Successes and Challenges 09:10 - Breakthrough in Robot Training 11:03 - Improving Performance 15:43 - Expanding Capabilities 17:34 - Robots in Unseen Environments 25:54 - Handling Open-Ended Prompts 29:36 - Evaluating Robot Performance 30:03 - Future Directions and Challenges 31:27 - Audience Q&A

Y Combinator

129,834 次观看 • 1 年前

Lutnick: You realize trillions of dollars of factories are going to be built in America. That's huge gdp. The factories being built in America are huge -- Brennan: That takes years, and you said that robots are going to fill those jobs. So those aren't union worker jobs. Lutnick: it’s automated factories.. The key is who is going to build and operate the factories Brennan: You said robots

Lutnick: You realize trillions of dollars of factories are going to be built in America. That's huge gdp. The factories being built in America are huge -- Brennan: That takes years, and you said that robots are going to fill those jobs. So those aren't union worker jobs. Lutnick: it’s automated factories.. The key is who is going to build and operate the factories Brennan: You said robots

Acyn

3,100,486 次观看 • 1 年前

Hello, Stockholm. 🇸🇪 From the podcasts we discover to the robots working in factories, the Nordics are building with NVIDIA. Spotify is making podcast recommendations more efficient. ABB Robotics is helping engineers train and test robots in simulation. And startups are adapting open models for work ranging from Swedish transcription to patent research. Behind these applications are investments in computing power and a community of researchers, entrepreneurs, and businesses bringing new ideas to life. Explore the Nordics’ technology ecosystem ahead of GTC Berlin:

Hello, Stockholm. 🇸🇪 From the podcasts we discover to the robots working in factories, the Nordics are building with NVIDIA. Spotify is making podcast recommendations more efficient. ABB Robotics is helping engineers train and test robots in simulation. And startups are adapting open models for work ranging from Swedish transcription to patent research. Behind these applications are investments in computing power and a community of researchers, entrepreneurs, and businesses bringing new ideas to life. Explore the Nordics’ technology ecosystem ahead of GTC Berlin:

NVIDIA

42,538 次观看 • 22 天前

This is why the West can’t stay without Africa. We have everything, and we need to start building factories and industries.

This is why the West can’t stay without Africa. We have everything, and we need to start building factories and industries.

Typical African

19,712 次观看 • 5 个月前

The rise of AI factories marks a new industrial era that is built on efficiency, intelligence, and sustainability. How we build them will shape economies, industries, and the future of the planet.

The rise of AI factories marks a new industrial era that is built on efficiency, intelligence, and sustainability. How we build them will shape economies, industries, and the future of the planet.

NVIDIA Newsroom

56,120 次观看 • 1 年前

Nox Metals (NOX METALS) builds fully automated factories that cut and ship metal fast. They use software and robotics to get factories cheaper metal, delivered in days. Congrats on the launch, Zane Hengsperger!

Nox Metals (NOX METALS) builds fully automated factories that cut and ship metal fast. They use software and robotics to get factories cheaper metal, delivered in days. Congrats on the launch, Zane Hengsperger!

Y Combinator

141,603 次观看 • 1 年前

We all need to keep up the pressure and demand that Trump follows the law and gets SNAP benefits to the millions of families and kids that are counting on them to put food on the table.

We all need to keep up the pressure and demand that Trump follows the law and gets SNAP benefits to the millions of families and kids that are counting on them to put food on the table.

Senator Patty Murray

23,859 次观看 • 1 年前

How do you know which robots are ready for the job? Robotics companies test their models on different tasks and datasets. Lightspeed Partner Amber Yang sees a need for better ways to compare them, especially when customers are making decisions about safety. She maps the robotics market by how general or specialized companies’ models and hardware are. As models become more adaptable, Amber sees an opportunity for companies that specialize in deployment. She calls them “DeployCos”: businesses that could pair existing models with different kinds of embodiments to handle work in places like factories or restaurants. 0:00 Why Older Robotics Companies Struggled 0:40 Mapping The Robotics Landscape Across Two Axes 1:16 Model Layer Breakthroughs From VLAs To World Action Models 4:20 One Brain Many Bodies And The Value Of Specific Models 5:27 In-Context Learning And How Robotics Companies Collect Training Data 7:40 Hardware Developments And Dependence On China 9:24 Specific Model Specific Embodiment Quadrant 13:05 The Deploy Co Angle And Open Router For Robotics 13:55 General Model Multiple Bodies Quadrant 15:50 General Model Specific Embodiment Quadrant 17:24 Where Things May Be Commoditized And Infrastructure Opportunities 20:45 Robotics Observability And Evaluation 22:33 Learning From Autonomous Vehicles Claire Zau

How do you know which robots are ready for the job? Robotics companies test their models on different tasks and datasets. Lightspeed Partner Amber Yang sees a need for better ways to compare them, especially when customers are making decisions about safety. She maps the robotics market by how general or specialized companies’ models and hardware are. As models become more adaptable, Amber sees an opportunity for companies that specialize in deployment. She calls them “DeployCos”: businesses that could pair existing models with different kinds of embodiments to handle work in places like factories or restaurants. 0:00 Why Older Robotics Companies Struggled 0:40 Mapping The Robotics Landscape Across Two Axes 1:16 Model Layer Breakthroughs From VLAs To World Action Models 4:20 One Brain Many Bodies And The Value Of Specific Models 5:27 In-Context Learning And How Robotics Companies Collect Training Data 7:40 Hardware Developments And Dependence On China 9:24 Specific Model Specific Embodiment Quadrant 13:05 The Deploy Co Angle And Open Router For Robotics 13:55 General Model Multiple Bodies Quadrant 15:50 General Model Specific Embodiment Quadrant 17:24 Where Things May Be Commoditized And Infrastructure Opportunities 20:45 Robotics Observability And Evaluation 22:33 Learning From Autonomous Vehicles Claire Zau

Lightspeed

51,942 次观看 • 1 个月前