America has been extremely worried about the loss of manufacturing to China. Seduced by subsidies, cheap labor, lax regulations, and a rigged currency, American industry has made a beeline to China.

But the tide may soon turn.

New technologies will likely cause the same hollowing out of China’s manufacturing industry over the next two decades that the U.S experienced over the past twenty years. That’s right. America is destined to once again gain its supremacy in manufacturing, and it will soon be China’s turn to worry.

China’s largest hi-tech product manufacturer Taiwan-based Foxconn Technology Group, made waves last August when it announced plans to install one million robots within three years to do the work that its workers presently do. These robots will perform repetitive, mechanical tasks to produce the circuit boards that go in many of the world’s most popular consumer gadgets. But even these robots and circuit boards will soon be obsolete.

As my colleague Neil Jacobstein, who co-chairs the Artificial Intelligence (AI) and Robotics program at Singularity University explains, there are three exponentially accelerating technologies—artificial intelligence, robotics, and digital manufacturing—that will reshape the competitive landscape for manufacturing. Specifically, these technologies will make manufacturing more creative, less expensive, more local and more personal.

AI is software that makes computers do things that, if humans did them, we would call them intelligent. This is the technology that IBM’s Deep Blue computer used to beat chess grandmaster Garry Kasparov in 1997, and that enabled IBM’s Watson to beat TV-show Jeopardy champions in 2011. AI is what powers the self-driving car that Google is developing and Apple’s Siri voice-recognition software. As a field, AI is now over 50 years old. People thought AI was dead after all the hype it generated in the ‘80s and failed to deliver. But it is fulfilling its potential now.

AI technologies will find their way into manufacturing and make it “personal,” but the technical challenge is simplifying the design process for products that we want to “manufacture” at home, says Jacobstein. Computer-aided-design companies like Autodesk are actively working to make what they call the Imagine, Design, Create-process much easier for mere mortals to perform. These will empower millions more people to join the ranks of the creator economy, where mass production is replaced by personalized production, and people are empowered to specify new products, design, test, and build them.

There is also robotics. Robots used to be a lot like the computers of the 1960s—extremely expensive, huge, slow, finicky, and difficult to program. However, Jacobstein claims that robots are currently in the midst of a revolution in cost and ease of use not unlike the transition from mainframes to personal computers and smart phones. Ten years ago, people would have laughed at the notion that robot vacuum cleaners like the Roomba from iRobot would be in over 6 million homes.

Today, robots are moving into multiple applications, from surveillance and telepresence, to surgery and manufacturing. Eventually, robots will become cheaper than human labor. They don’t sleep, take days off, get distracted, or (currently) demand ever higher wages. American robots will soon compete directly with Chinese labor, allowing the U.S. to produce many types of products to specifications in our own homes. Don’t expect these manufacturing robots to necessarily look like a conventional one- or two-armed robot. Rather, except them to be be embedded in a desktop manufacturing unit that becomes the basis of a young entrepreneur’s business.

Digital manufacturing is another piece of the competitive manufacturing puzzle. It refers to a spectrum of capabilities that include the ability to imagine new products and test them virtually using design checkers and simulators, specify the design of three-dimensional objects in computer software and send that design to a 3D printer. These materials printers can render the design in plastic, composites, or metal in a matter of minutes or hours, depending on the size and complexity of the design.

Today, simple desktop 3D printers produce relatively crude objects. These 3D kits sell for between $500 and $1000. Imagine a toothpaste tube of plastic or other material held vertically in an X-Y plotter that squirts out thin layers of tiny dots of material that are built up, layer by layer, to produce a 3D replica of the design in the computer. The resolution of 3D printers varies with cost, but relatively inexpensive machines have a resolution of 100 micrometers. These manufacturing machines are evolving rapidly, dropping in price and increasing in capabilities. By the mid-2020s, we will develop advanced nanotechnology or molecular manufacturing which will allow us to program molecules inexpensively, with atomic precision, according to Jacobstein. Molecular manufacturing will do for our relationship with molecules and matter what the computer did for our relationship with bits and information. Specifically, it will make the ability to program molecules into precise 3D objects inexpensive and ubiquitous.

What happens when you combine AI, robotics, and digital manufacturing? A manufacturing revolution, that will enable U.S. entrepreneurs to “set up shop” locally, and create a wide variety of products. As Kinko’s is for 2D digital printing on paper, we will have shared public manufacturing facilities like TechShop where you can print your 3D products. How is China going to compete with that?