Author: Syed Armani
Compiled by: Felix, PANews
AI is no longer confined to screens and software. As AI merges with robotics, machines are gaining the ability to perceive the world, interpret changing conditions, and act in real time. This shift toward intelligent physical systems (i.e., physical AI) is beginning to reshape industries and is expected to impact everyday family life as the technology matures.
Innovation in robotics is surging at an unprecedented pace. Figure recently launched the Figure 03 humanoid robot, designed for home and commercial applications. It can perform some household chores, such as folding clothes and loading the dishwasher, but it's not perfect yet. Tesla is running the Optimus humanoid robot in a limited internal pilot project on its factory floor. Autonomous drones and legged robots are increasingly being used for hazardous inspection tasks. Meanwhile, Unitree Robotics and haptic technologies like FlexiTac are working to enable robots to navigate cluttered home environments, ensuring safety around pets and children and assisting with everyday tasks. Once fully developed, intelligent robots will focus on general intelligence and contextual awareness, such as recognizing a spilled glass of water that needs to be cleaned up without explicit instructions.
Investors are pouring significant sums into the technology stack that promises to underpin next-generation robotic hardware. In January 2026, Skild AI raised $1.4 billion in its Series C funding round, valuing the company at $14 billion, to expand its general-purpose robotics foundation model; while Figure AI raised over $1 billion in its 2025 Series C funding round, valuing the company at $39 billion post-money, to expand human manufacturing capabilities and industrial deployments. Apptronik increased its Series A funding to $935 million, and NEURA Robotics added €120 million to its Series B funding round. These developments highlight a growing consensus: physical AI is becoming the strategic foundation for consumer and industrial robotics.
Has the tipping point for the widespread adoption of intelligent robots arrived?
The current accelerated development in this field is the result of the convergence of multiple technologies. For decades, the various modules that make up intelligent robots were developed independently, such as advanced AI algorithms, high-fidelity sensors, robotic arms, and real-time control systems. Only recently have these modules begun to merge, enabling robots to effectively perceive, reason, and act in real-world environments. The following are the key factors driving this "inflection point in robotics":
Economic factors: Hardware has finally been commoditized. In the past, robots were expensive because each component was custom-made. Now, they benefit from the supply chains of consumer electronics and electric vehicles.
Actuators: Actuators for high-torque humanoid robots have historically been expensive, with each joint typically costing over $1,000 in small-batch industrial systems. However, new vertically integrated designs from companies like Tesla and Unitree Robotics are reducing the cost of some actuator components to a few hundred dollars.
Sensors: Over the past decade, the cost of LiDAR and depth cameras has dropped dramatically. High-end devices that once cost around $10,000 now cost only a few hundred dollars. This is thanks to advancements in solid-state design, mass production, and applications in the automotive and mobile device sectors.
Batteries: Global investment in electric vehicles has reduced the cost of high-density lithium-ion batteries and improved their reliability, enabling many robots to run for 2-4 hours on a single charge.
Edge computing: Robots must process information locally because real-time control tasks such as balancing or grasping objects do not allow for network latency. Chips like NVIDIA's Jetson Thor are designed to run AI inference on-board, processing multiple sensor data streams simultaneously. This allows robots to process and track their environment locally, responding quickly to changing conditions without relying on a network connection.
Breakthrough in AI Models: This is the biggest change. It's a shift from "if/then" programming to "World Models." A World Model is an AI model that learns how the real world works by watching videos. Instead of programming a robot to "turn a doorknob," it's shown 10,000 videos of doors opening. The AI simply observes the videos to build mental models of how physics works, develops physical intuition, and mentally simulates scenarios before taking action. Google Deepmind Genie 3 and NVIDIA Cosmos are examples of this new type of World Model.
As machines become more intelligent, costs continue to fall. For example, the Noetix Bumi robot (priced at $1,400) is now roughly the same price as an iPhone 17 Pro Max. Decreasing hardware costs, improved AI chip performance, and enhanced world-modeling capabilities are all contributing to making intelligent robots more accessible to the public and expanding research and development from cutting-edge technology labs to a wider range of fields.
If the "ChatGPT moment" in robotics arrives soon, we will likely first see applications in the industrial and logistics sectors, followed by truly humanoid home robots. While many challenges remain before intelligent robots become widespread, rational optimists recognize that current trends point to a future where the widespread use of intelligent robots is increasingly likely.
Major software breakthroughs are often accompanied by hardware breakthroughs. The emergence of Instagram and TikTok is a prime example of how essential hardware can drive growth. If intelligent robotic hardware becomes widespread in the near future, an interesting question arises: will robotic applications be the next big thing?
What challenges are currently hindering this development momentum?
Robot training data: This is the biggest bottleneck facing the development of general-purpose intelligent robots. Unlike text-based AI that can crawl the entire internet, robots need real-world experience, such as sensory perception, maintaining balance, and interacting with objects. Collecting this type of data is slow, expensive, and extremely labor-intensive.
The "physical" issue: Watching videos cannot fully teach a robot how to manipulate objects or move safely; it must experience forces and contact firsthand. Remote operation, where a human guides the robot in real time, can capture both intent and force simultaneously, making it the best standard for data collection. Generating hundreds of hours of high-quality data requires the operator's full presence, and its scalability is far inferior to digital data collection.
The gap between simulation and reality: Simulation can generate large amounts of data at low cost, but because physical phenomena are not modeled or the environment is unpredictable, robots often fail to transfer skills to the real world.
On-chain machine economy
The combination of blockchain and robotics offers a practical solution to the challenges currently facing robotics technology. Token incentive mechanisms can help coordinate millions of robots and reward contributors to remotely operating devices or sensor data. Every interaction becomes a valuable data asset, building a rapidly growing, community-owned robot dataset far exceeding the scale of any single company.
Tokenization of data collection
Robotics data is extremely valuable, but real-world sensing and interaction data is scarce. Large companies collect massive amounts of driving and industrial data through their fleets, giving independent developers an unparalleled scale advantage.
Decentralized physical AI allows users to remotely control robots or contribute sensor data and earn token incentives. Decentralized networks can coordinate thousands of enthusiasts globally to help robots navigate complex surfaces or special environments. Contributors can upload data and receive rewards. Although these platforms are still in their early stages, they foreshadow a future where robot data can be shared more widely, weakening the monopoly of a few large companies.
Robots as economic agents
In the "Robots as a Service" model, the intelligent robot itself can become a "tokenized" asset. Each robot (or right to use) can be represented by digital tokens, allowing multiple users to own or lease it. Service fees paid to the robot can be deposited directly into the robot's wallet via tokens or stablecoins. This setup enables self-generated revenue: the robot earns money by working, covers its own operating costs, and automatically distributes profits to token holders. Essentially, this is a Web3 protocol that transforms robots into programmable, self-sufficient service providers with transparent and traceable revenue.
Physical AI Market Landscape
With the advancement of next-generation intelligent machine learning and the understanding of the complex realities of the three-dimensional world, the boundaries between digital intelligence and physical behavior are disappearing.
At the heart of this revolution are AI models. Developed by Physical Intelligence and Skild AI, sophisticated “brains” transcend static code, providing general intelligence across various physical forms. These models allow robots to treat agility and mobility as software problems, enabling a single, unified “brain” to adapt to multiple robotic bodies. This layer of intelligence is supported by simulation platforms and data pipelines (such as those provided by Zeromatter), allowing systems to be securely trained in virtual environments before being deployed to the real world.
Alongside the development of robotic brains, decentralized physical AI is emerging. For example, the decentralized infrastructure network Fabric Protocol provides autonomous robots with on-chain identities and cryptographic wallets, and uses cryptography to verify machine operations. Companies like Auki, Peaq, and IoTeX are building a “machine economy” where robots can share 3D maps, verify data, and trade autonomously. This decentralized approach ensures that the coordination layer is not controlled by a single enterprise.
In the industrial sector, Bedrock Robotics' autonomous construction equipment and Mytra's warehouse automation are redefining the workforce, while ANYbotics handles routine maintenance tasks in hazardous environments. Meanwhile, with advancements from Figure and Unitree Robotics, breakthroughs in home assistants are just around the corner in the consumer market.
2030 Outlook
From a rationally optimistic perspective, the resurgence of robotics has already arrived. Four unstoppable forces are converging: hardware costs are plummeting, AI model intelligence is constantly rising, edge computing chips are providing unprecedented computing power, and global industrial workers are poised to solve data challenges. By 2030, this synergy will drive the penetration of physical AI into every corner of the world, from autonomous agriculture to high-risk fields such as firefighting and elderly care.
History shows that transformative software innovations typically occur after hardware has stabilized. Perhaps we are entering an era of "smart leasing," where standardized humanoid robots will run standard operating systems and integrate app stores. Just as with the previous smartphone revolution, the next few years will be defined by "robot app stores," where users won't need to buy dedicated equipment but will instead subscribe to the robot's skills. In this model, value shifts from the machine itself to the specific "skills" it can perform. You won't need to buy a dedicated French tutor robot; simply download a "French skills app" to your general-purpose humanoid robot, and it will become your French teacher. By 2030, for the wealthy, the preferred holiday gift will no longer be a flagship foldable phone, but a smart assistant that can truly help manage household chores.
This prediction is based on rational optimism. Although the road to the future is rarely smooth, the convergence of various technologies is foreshadowing a profound revolution in machine technology.
Related reading:When robots learn to think, earn money, and collaborate: Analyzing 15 types of robot technologies and application cases.
