What Is the ARMS Framework? Agents, Robots, Materials Science, Explained
ARMS stands for Agents, Robots, Materials Science — the Optimus technology thesis created by Brad Hart. It maps three horizons — Agents (the NOW), Robots (the NEAR, a 1–2 year horizon), and Materials Science (the NEXT) — as one compounding trajectory from software to atoms, and tells a founder which layer to build on and in what order.
Most AI frameworks answer a tooling question: which model, which vendor, which stack. ARMS answers an architectural one: where is the world actually going, and where should a business owner place their weight? That distinction matters, because tooling decisions get obsoleted every quarter. A thesis, if it's right, compounds for a decade.
Why does the ARMS framework exist?
Because the information economy is compressing. AI puts a world-class expert in every pocket at a fraction of the old cost — the ARMS roadmap puts it at 1/100. If your business sells knowledge, advice, or expertise, that product is being commoditized in real time. And physical services and goods businesses aren't safe either — they just have a couple more years before agent-controlled robotics reaches their loading dock.
The usual escape routes are traps. Go down-market and you're in a race to zero against software that works for free. Go enterprise and you're in eighteen-month sales cycles against incumbents. ARMS proposes a third move: go deeper in the value chain, or wider in leverage — and it names the three layers where that depth and leverage live.
What are the three horizons of ARMS?
A — Agents: The NOW
Any work a human does with software, an agent can do — autonomously, faster, and cheaper. This is not a prediction; it's what gets deployed today via the FAST framework (Factory of Agents with Skills and Tools). The claim ARMS makes for this layer: 10–100× output from the same team, full-stack automation across ops, sales, marketing, support, and finance — scale without headcount.
R — Robots: The NEAR
Agents are software minds. Robots give them hands. The thesis puts practical, at-scale robotics deployment on a 1–2 year horizon: warehouses, manufacturing, logistics — agent-controlled physical systems optimizing a business end-to-end, from decision to delivery. The brain already exists. The body is catching up fast.
MS — Materials Science: The NEXT
Once you unlock energy and compute, you unlock everything: new materials, new chemistry, new biology — and the IP to own the breakthroughs. This is the layer ARMS calls the NEXT, and it's already under construction through AIMS (AI Materials Science) research: 78 patents filed, 41 papers published, roughly 1,275 claims across 40 scientific domains.
What makes ARMS compound instead of just grow?
The horizons aren't three separate bets. They're a loop, and each turn funds the next:
- Agents → virtually free labor. Radical efficiency frees capacity and capital to deploy into the next layer.
- Robots → new markets. Physical automation opens revenue streams no software-only company can touch.
- Materials science → new value. IP, licensing, and breakthrough products create categories that didn't exist.
A business that masters Horizon 1 doesn't just get more efficient — it earns the free cash flow and the operating capability to enter Horizon 2 early, and the compounding repeats. That's the difference between linear growth and a trajectory.
How does a founder actually use ARMS?
ARMS resolves into four business plays, each drawing on a different mix of the three layers:
| Play | The move |
|---|---|
| Optimize & Flip | Deploy agents and robots into existing businesses — starting with yours. Boost efficiency, raise valuations, exit at a premium. |
| Roll-Ups & Acquisitions | Absorb businesses, drop in the AI stack, multiply their output with the same team. |
| Materials Science Ventures | Go deeper — licensing, IP, and new products built on breakthrough research. |
| Consulting & Deployment | Help other companies deploy ARMS capabilities and capture revenue across every layer. |
Whichever play fits your situation, the sequencing is the same: master agents first. The practical entry point is covered step-by-step in how to deploy AI agents in your business.
Where does ARMS sit inside Optimus?
ARMS is one of the eight Optimus Frameworks, and it's the thesis layer — the one that answers "where is this all going?" The others answer the questions that follow: OSLO handles prioritization (what to work on, in what order), FAST is the engine that delivers Horizon 1 today, and AIMS is the research frontier behind Horizon 3. The full map lives at optimusframeworks.com.
If you're weighing whether the timing is right, two companion reads: what comes after AI agents lays out the horizon sequence in detail, and is it too late to get into AI answers the question most founders are actually asking.
FAQ
Is ARMS a product or a thesis?
A thesis. ARMS is the map, not the vehicle. It tells you where technology is going — agents now, robots near, materials science next — so you can decide which layer to build on and in what order. The vehicles are the other Optimus frameworks: FAST for deploying agents, OSLO for prioritizing the work, AIMS for the research frontier.
What does ARMS stand for?
Agents, Robots, Materials Science. The A is Horizon 1 (the NOW), R is Horizon 2 (the NEAR, a 1–2 year horizon), and MS is Horizon 3 (the NEXT). Each horizon compounds the value of the one before it.
Who is the ARMS framework for?
Founders who own a real business — typically in the $5–50M range — and need a straight answer to the question "AI is clearly rewriting my industry; what do I actually build, and in what order?" ARMS is the ordering: master agents first, integrate robotics as it matures, and position for the materials science layer.
How does ARMS relate to FAST, OSLO, and AIMS?
ARMS is the thesis — where the world is going. OSLO is the prioritization framework — what to work on, in what order. FAST is the engine — the Factory of Agents with Skills and Tools that delivers Horizon 1 today. AIMS is the research frontier that delivers Horizon 3. They're four of the eight Optimus Frameworks created by Brad Hart.