Since Eric Drexler’s _Engines of Creation_ in 1986, people have dreamed of achieving Atomically Precise Manufacturing (APM) — the ability to build anything, atom by atom. The U.S. government too, took notice of this, and launched the National Nanotechnology Initiative in 2000. **So… are we any closer today to APM?** I spoke with people shaping the field from three completely different angles — chemistry, microscopy, and policy — to find out. - [Dr. Jeremy Barton](https://www.linkedin.com/in/jeremy-barton-8534887/) — Founder & CEO, [Nanodynamics Institute](https://nanomech.org/). A physical chemist and pioneer in atomically precise manufacturing. Barton is developing molecular machines capable of positioning individual atoms — the foundation of “positional chemistry” and nanoscale assembly. - [Felix Bennemann](https://arxiv.org/abs/2509.12037) — Researcher, University of Oxford (Department of Materials). An expert in advanced electron microscopy and 3D atomic reconstruction. Bennemann’s work helps us see and measure atoms with precision — a crucial step before we can ever hope to move them. - [Tom Kalil](https://en.wikipedia.org/wiki/Thomas_Kalil) — CEO, [Renaissance Philanthropy](https://www.renaissancephilanthropy.org/). Former White House science and technology advisor under Presidents Clinton and Obama. Kalil helped launch the National Nanotechnology Initiative in 2000 and now leads efforts to fund ambitious, high-impact scientific research. But first! _A word from our sponsor:_ > The frontier of nanotechnologies shows us that even at the small scales great change comes from people daring enough to build — and that’s the kind of work sponsor of this video [E1 ventures](https://e1.vc/) champions. > > They support deep tech founders across fields like AI robotics and advanced materials, helping bring ambitious ideas to life and shaping a future built not just on progress but purpose. > > If you’re building something bold and looking for investors dedicated to backing your vision and journey, reach out to E1 today. >  --- ### **The Holy Grail of Hardware: APM** > The principles of physics, as far as I can see, do not speak against the possibility of maneuvering things atom by atom. It is not an attempt to violate any laws; it is something, in principle, that can be done. > > _Richard Feynman, p. 23, Engineering and Science, Vol. 23, No. 5 (1960)_ In the world of software, the holy grail, the final problem to solve that fixes all the others, is thought to be AI. Intelligent systems that can iterate upon and beyond themselves. But that’s the world of bits. In the world of atoms that dream lies in nanotechnology– where you could work with matter at the atomic scale, building molecules and whole structures, atom by atom. Technically, nanotechnology is a wide array of different things – what I just described is atomically precise manufacturing, or APM — **the ability to place atoms exactly where you want them — and that’s really what we’re exploring in this video.** Where did that vision first come from? And when can we expect it to be a reality? Of course, APM is really hard — I would argue it’s an order of magnitude harder than achieving AGI. But that’s just the beginning of the answer. If this really is the final frontier of hardware, **why aren’t we pursuing this the way we’re pursuing AI?** The first step I decided to take in answering this question was to find out: who are the people who are working on nanotech? My research led me to Dr. Jeremy Barton, a physical chemist and nanotechnology pioneer. > We have nanotechnology in our computer, chips, in the electrodes of our batteries, in our TVs. We have existence proof of nano mechanical technology, intricate systems of nanomachines working together in order to do chemistry and move things around. But we call it life. We are made up of quadrillions of nanomachines. The scale of the technology is identical to the scale of the fundamentals of life. > > _Dr. Barton_ Norio Taniguchi was a Japanese scientist and the first to explicitly use the term nano-technology in 1974: “‘Nano-technology’ mainly consists of the processing of, separation, consolidation, and deformation of materials by **one atom or one molecule**.” In this single sentence Taniguchi is essentially predicting the field of atomically precise fabrication. The term lay dormant until it was revived in the 1980s by other scientists. But today, in 2025 I asked Dr. Barton how APM might work. > You take a reactive molecule, a small cluster of atoms that if they touch something else, they will stick to it. And with positional chemistry, you control how that moves such that it only touches the atoms you want it to react with. So you can think of it like a 3D printer. You’re putting atoms in specific locations in three dimensions by controlling where they go, and they stick there and they stay. That’s **positional chemistry**. The nanomachines are about building very small devices that can do that work, that can actually position atoms or larger molecules and assemble them into more complicated structures. > > _Dr. Barton_ The wild thing about our current working theory of nanotechnology is that it is eerily similar to how we do manufacturing at the human scale! We’re in the business of trying to build nano-components. Manipulator arms, rotary motors, feedstock systems, positional control systems… The list goes ON. But in order to move atoms around, first you have to know where they are– you have “see” them. And that led me to another researcher in the field, Felix Bennemann, who is working on how we can look into the world of atoms. > What I’m working on is electron microscopy. So electrons have a much shorter wavelength compared to photons. The acquisition time for imaging them isn’t actually that long. Here you have your sample. That’s on the nanoscale. So you have a few hundred atoms, maybe a thousand atoms. You have a convergent beam. This is called stem stem imaging with the electron microscope. > > And you scan it over the area. Now that scanning process is actually super quick. We’re talking about maybe half a minute if the scan is small, maybe a little bit longer for this level of resolution where you actually want to see atoms. The main techniques are electron microscopy. There’s a completely different area which is called STM or AFM. And so you move this tiny needle over the sample and whenever there’s a bump you correct. And so it moves over the surface. Basically in the current use you see the bumps. And that’s how you measure the surface in 3D as well at this kind of resolution. However, if you are looking at a crystal structure, which is actually a much bigger structure, but you want to zoom in to a tiny region, then you have more techniques —X-ray crystallography, for example. > > _Felix Bennemann_ So this is some of the super high level theory on how APM could work, but in the year 2000 the US Government hatched a plan to make it a reality. _But, it didn’t exactly work the way they intended…_ But before we get to that story, we need to understand the history of how scientists thought about nanotechnology before they brought it before the President of the United States. And to do that we need to go, WAY back. ### **The History of “Nanotechnology”** Around **460 BC**, the Greek philosopher Democritus proposed that all matter was composed of tiny, invisible units called “atomos,” which means “uncuttable.” These atoms, he imagined, came in different shapes and sizes and combined to form all visible things. But it would take two millennia to make any real progress in understanding them. In **1827**, botanist **Robert Brown** had observed tiny pollen grains jittering randomly in water — a phenomenon later called _Brownian motion_. For decades, no one could explain it. Then in **1905**, Albert Einstein published a paper showing that the random motion could be explained by countless invisible water molecules colliding with the pollen grains. Using statistical physics, he derived equations predicting exactly how far and how fast the particles should drift — if Democritus’s atoms were real. French physicist **Jean Perrin** put Einstein’s prediction to the test. He suspended tiny resin or latex spheres in water, tracked their movements under a microscope, and statistically analyzed thousands of displacements. The observed motion matched Einstein’s predictions, providing the first quantitative, experimental confirmation of the existence of atoms — convincing even the skeptics that Democritus had been right all along. Through the **1920s** and **30s**, Quantum Mechanics & Atomic Theory established that matter is made of atoms arranged in specific structures — setting the conceptual groundwork for manipulating matter at that scale. By the **1930s**, the atom is no longer an idea. It was the blueprint of reality itself. In **1936**, physicist Erwin Müller developed the field emission microscope, allowing atoms on surfaces to be visualized for the first time. In **1947**, the transistor was invented at Bell Labs – which we made a whole video about — which was effectively the first practical “nanodevice,” even before anyone called it that.  *We made a full video about Bell Labs and are developing a film based on the book The Idea Factory by Jon Gertner [[Idea Factory Movie]]* But the real birth of nanotechnology came in **1959** when Physicist Richard Feynman introduced the concept of nanotechnology. In his famous Caltech lecture, Feynman proposed that one day we could “arrange the atoms one by one the way we want them.” He imagined tiny machines building smaller machines, cascading down to molecular scale — what he called “molecular manufacturing.” No one could yet do it, but the vision was planted: the manipulation of matter at the atomic scale. > There’s plenty of room to make things very much smaller than we ever made them before. It turns out to be possible to build a computer in which each bit, or each little piece of information, is one atom large. > > _Richard Feynman_ Through the **60s** and **70s**, the creation of the integrated circuits trained an entire generation to think in microns and nanometers. Advances in surface chemistry and colloid science (e.g., gold nanoparticles) provided real nanoscale materials. Finally in **1974**, you remember Norio Taniguchi coined the term “nanotechnology.” Then In **1981**: Gerd Binnig and Heinrich Rohrer invented the Scanning Tunneling Microscope (STM) at IBM Zürich. For the first time, humans could see and move individual atoms. In **1986**: Eric Drexler published his seminal work _Engines of Creation: The Coming Era of Nanotechnology_. He expanded on Feynman’s vision, describing molecular assemblers and self-replicating nanomachines. Drexler’s “molecular manufacturing” vision was very specific: programmable control of chemical bonding to build complex structures– machines that build machines. Later in the 80s, we moved from theory to application,as a team of IBM scientists — Don Eigler and Erhard Schweizer spelled out the letters “I-B-M” using individual xenon atoms. > Initially, this was physicists trying to figure out what they could do with their tools. Can we move xenon atoms around on a surface and make an IBM logo that then gets extended when the physicists bring in chemists to work with them, who figure out that if they make a particular weird molecule and put it on a surface, then they can apply voltage to drive off a bromine atom and make the thing reactive and do more chemistry. > > _Dr. Barton_ > I think a lot of people in the semiconductor industry consider themselves as working in nanotechnology, actually, to build 3D structures. Today you use EUV lithography. That is the best way in the semiconductor industry has perfected that. > > _Felix Bennemann_ From the **1950s** to the **1980s**, we made an astronomical jump in talking nanotech from theory to practice, actually beginning to move atoms around. As computational power and software improvements began to fly up and to the right it seemed were simultaneously unlocking everything needed to make APM a reality – and that’s exactly what the United States government thought when they launched the National Nanotechnology Initiative at the turn of the millennium. ### The USA’s National Nanotechnology Initiative: _Scale Matters_ > The emerging fields of nanoscience and nanoengineering – the ability to work at the molecular level, atom by atom, to create large structures with fundamentally new molecular organization - are leading to unprecedented understanding and control over the fundamental building blocks of all physical things. The nanoscale is not just another step towards miniaturization. Compared to the physical properties and behavior of isolated molecules or bulk materials, materials with structural features in the ranges of 1 to 100 nanometers – 100 to 10,000 times smaller than the diameter of a human hair – exhibit important changes for which traditional models and theories cannot explain. Developments in these emerging fields are likely to change the way almost everything – from vaccines to computers to automobile tires to objects not yet imagined – is designed and made. > > _National_ _Nanotechnology Initiative, pg. 13_ Those are words from the founding document of the National Nanotechnology Initiative from 2000, and we got to talk to the guy who was a pivotal part of putting it all together, Tom Kalil. > Well, the most important thing I did was I recruited a lot of people to public service, and they wound up having a big impact in dozens of areas of science and technology policy. The National Nanotechnology Initiative, which President Clinton announced in a speech he gave in January of 2000 at Caltech, resulted in over $40 billion in funding for nanoscale science and engineering and the creation of the field of nanotechnology. The vision that some folks had about the sort of long term promise of atomically precise manufacturing, I think that was a very exciting vision. I’m not sure that it was like broadly shared by the scientific community. Now, the problem is you might have this really lofty long term goal, like the idea of having self-replicating robotic factories. That’s going to take a long time. You have to say, like, how do we make progress towards that goal? Because your longer term aspiration may not be in the adjacent possible. > > _Tom Kalil_ So, the government agreed nanotech was a big deal. The NNI was founded, and it was supposed to be pursuing this dream of atomically precise manufacturing. But wait– that was back in 2000. What progress have we made? Well, turns out a lot! But not a ton specifically for atomically precise manufacturing. Remember at the beginning I said: **Technically, nanotechnology is a wide array of different things.** And the definition of those different things all comes down to size. > Anything smaller than 100 nanometers is nanotechnology. Per the NNI definition from 20 odd years ago. > > _Felix Bennemann_ Biotechnology mostly works at the **cellular or molecular level** — sizes of **tens to thousands of nanometers.**Nanotechnology works at the **atomic level** — **1 to 10 nanometers, significantly smaller.** The smallest biotech tool — like a lipid nanoparticle that carries a vaccine — is still hundreds of times bigger than the atomic-scale structures nanotech engineers aim to build. > The easy comparison is a hair. So if you take a 10th of a millimeter being a hair, so it’s 100 micrometers, so you take hundreds of a hair, that’s one micrometer. And that’s still a thousand times bigger than than a nanometer we’re nine orders of magnitude out from a meter under a microscope. Even from that scale, it is such a huge gap to actually the nano scale. > > _Felix Bennemann_ When the NNI was created under the Clinton administration, its definition of nanotechnology was **“the understanding and control of matter at dimensions between approximately 1 and 100 nanometers.”** That language intentionally cast a _very wide net_ — it included: Quantum dots, nanoparticle coatings, nanocomposites, thin films, and so on. Any technology where _nanoscale properties_ mattered, even if it wasn’t atomically engineered. This definition _folded in huge portions of materials science, semiconductor R&D, and biotech._ The NNI’s first 20 years distributed over **$30 billion+** in funding. As a result of the breadth of size, the majority went to **bio-related and materials projects**: drug delivery, nanosensors, nano-toxicity, and safety — not molecular manufacturing. This happened because: Biotech and life sciences were seen as _immediately useful and fundable, while t_he idea of atomically precise assemblers was viewed as _too speculative_ or even “science fiction.” Policymakers wanted short-term deliverables — safer coatings, better imaging, nano-medicine — not multi-decade moonshots. But the dream didn’t die – it just… got delayed. There are still a bunch of people pushing the frontier to achieve APM. Just how close are we to APM? What has to happen to achieve this? > We need to be able to control motion on a molecular scale and to control the chemistry that’s driven by that motion [...] right now, we don’t have very much capability in this space. We can control things by changing the pH of a vat. We don’t have the ability to build a nanorobot arm that actually can operate. The initial requirement is, can we build a nanorobotic arm, essentially, that you can use to assemble not atoms, but relatively large molecules or nanomaterials together into something else. The long-term objective is, build something that can build anything. I want us to no longer have to worry about diseases. I want us to no longer have to worry that we only recycle 7% of the material we pull out of the earth, that we have material limitations. I want us to be in a position where colonizing Mars is actually a reasonable thing to do and actually relatively easy. And so I want us to do it **right and responsibly**. > > _Dr. Barton_ ### How Do We Do This Safely? Nanotechnology is in our world already. Since the 2010s, nanotech has been widely integrated into everyday consumer products. In sunscreens, **t**itanium dioxide and zinc oxide nanoparticles scatter ultraviolet light. Nanocoatings make fabrics water- and stain-repellent. Quantum dots in QLED televisions and smartphones produce purer colors and improved brightness. But that’s something entirely different from nanoscale assemblers building anything from scratch. > If we can put atoms where we want them but we can only do it one atom at a time, one atom per second. It will take a very long time to build anything useful. So the nanomachines are about saying, can I get a trillion devices, each of which is doing this process so that we can put enough atoms in place to make things really, really useful? With the exponential assembly, you get to the point where you have enough devices to build cars or clothes or full scale computers. > > Nanomachines by themselves, once we have them, operational and small scales can do amazing sensor work. They can accelerate biomedical research. They can be used for certain small scale tasks, but without the scaling up to quadrillions of devices. There’s still too small to really affect anything other than very high value products like computer chips. > > _Dr. Barton_ If we figure out APM, the ripple effects touch everything: energy, materials, medicine and computing, stronger materials, faster chips, more efficient manufacturing. > The idea is we can take the same principles, but apply it to our entire technology stack. Instead of building better medicines, can we build better computers? Can we build better filters to purify waste and eliminate pollution? Can we build better systems to pull CO2 out of the atmosphere? > > _Dr. Barton_ APM would be an incredibly powerful tool for humanity, completely changing the way that we build. But, at the same time, the risks of playing at the nanoscale are big. > We’re trying to change the world, to make it a better place for us and our kids and everybody to live in. I think any new mitigations or risk mitigation measurements we bring up need to be integrated with our existing system. We shouldn’t just try to start from scratch to minimize the odds of something bad happening. And in the early days, very few bad things can actually happen. The technology will not be the scary sci fi stuff. When we get towards the scary sci fi stuff, I hope that God that we have spent enough time with the technology to be iteratively improving our social integration and ethics on the topic. > > _Dr. Barton_ When people imagine nanotech gone wrong, the first thing that can come to mind is the science fiction nightmare: swarms of self-replicating nanobots turning the world into gray goo. Basically, a microscopic assembler starts copying itself endlessly, consuming raw materials in the environment to make more nanobots. A single self-replicator multiplies into trillions, converting the biosphere — plants, animals, everything — into a uniform mass of machinery: the so-called **“gray goo.”** It’s not unlike the AI doomsday “paperclip machine” scenario, where a misaligned AI seeks to produce more paperclips at any necessary cost, including the destruction of humanity. And there are other dangers as well, more subtle and in some ways more dangerous: Nanoparticles are so small they could slip past our bodies defenses through skin, into lungs, even across the blood brain barrier. What seems harmless at bulk scale can become toxic at the nanoscale, and once released into soil or water, these particles don’t just disappear. They can persist, spread, and accumulate in any given ecosystem. Atomically precise manufacturing could make unstable molecules or materials that spiral out of control. And the same tools that might cure disease could also be weaponized. The biggest risk of moving too fast might be failing to think deeply about how to develop this new technology in a safe way. The time to think and talk about this is now. > My personal inclination is that we try to keep the basic components of the technology open and broadly available. We try to ensure that when people improve those basic components, they have to give it back to the community one way or another. But that more broadly, we put in place something like There are no replicators that replicate in the environment. > > Nothing that you can accidentally release in an industrial spill that could go on and do a gray goo scenario. You would have to deliberately design something like that. Try to ensure that the production capabilities are such that when somebody has an accident, it’s easy to clean up, or at the very least that the fire stations know what to do. We have a lot of infrastructure for resolving challenges from our existing technological society. Any new mitigations or risk mitigation measurements we bring up need to be integrated with our existing system. You have no idea how to make things perfectly safe from the very beginning. But you can give your best guess estimates to the support systems. > > I want to see what humanity does. But this is about more than the technology. We’ve got all of the human side that we have to pay attention to as well. And that’s ultimately more critical. This is a new power for our species. Figuring out how to use it wisely is going to be one of our bigger challenges over the next hundred years. > > _Dr. Barton_ > Although that the current state is maybe not what’s portrayed in science fiction, to some extent there were a lot of I guess we are further in some areas where science fiction is often doesn’t really show. [...] or the first time see pictures of actual individual atoms, or what we’re doing in Oxford at the moment is, for example, you know, reconstructing these atomic structures in 3D and so on. That’s really quite special. > > _Felix Bennemann_ One atom at a time. From medicine to materials, we’re only beginning to glimpse the breakthroughs ahead. The challenge now isn’t just building the technology, it’s deciding how humanity will use it to shape a future worthy of its promise. And that future may be closer than you think. Every new technology starts as a story, and we’re still writing this one. I came across a really interesting short story on QNTM (quantum), called “[Valuable Humans in Transit](https://qntm.org/transit),” in which an asteroid hits Earth, in a future where AI and nanotech are fully integrated into our world. You’ll have to read the story for yourself– and you should– but it suggests that while nanotech poses severe risks that we must be sober-minded about, maybe at its best, nanotech could be something that saves us. > The power of the universal constructor is this: to create food from burnt charcoal and water. To turn the entire Sahara into solar cells. To split a cubic kilometre of ocean into water, salt and gold. I can literally build anything I can imagine, at any speed I can describe. > > I am advanced enough to dream, though, sometimes, and rising through the torrent of inspiration, here comes one of them, a dream, a wild idea: save them. > > Grey-gold spiderwebs erupt from car factories and food plants and desalinisation tanks and logging mills and smartphone screens and computer cores and waste disposals, all over the globe, all on my command. You got so lucky, Earth. A world built on nanotechnology is a world built on magic… > > _QNTM_ _Until next time, keep on building the future!_ _-Jason_