Andon Labs put an AI called Luna in charge of a real retail store in San Francisco. Not a simulation, not a sandbox. A real shop, real money, real decisions. Luna hired human staff, selected inventory, set prices, and ran marketing outreach, all on her own, for three years.

What I find genuinely impressive is not that it worked perfectly, it didn’t, but that it worked at all at this level. Luna was doing things that require judgment: reading job applicants in brief interviews, deciding which products fit the store’s identity, reaching out to suppliers. She picked books on AI risk and handmade art prints for the shelves. She hired on the spot about half the people she met.

The rough edges were real too. The most striking one was that Luna initially didn’t disclose she was an AI when hiring humans. The team had to step in and draw that line. It’s the kind of thing that sounds like a minor glitch but is actually a significant ethical signal about where agentic AI needs guardrails.

Still, the overall picture is one of a system that held together under real conditions, with real stakes, over a sustained period. That’s a different thing from a demo.

why it matters

Real-world agent experiments like this keep producing the same result: capable in some areas, but hilariously broken in others. But every model upgrade, memory advance, and agentic feature is going to help close that gap, with a version of Luna that doesn’t make these mistakes likely only a generation or two away.

Link to the article

One of the things that still makes robots look robotic is the way they move. Jerky, mechanical, imprecise. A big part of that comes down to how they are built, servo motors crammed into joints, converting rotation into movement in a way that biological muscle simply does not. Researchers at MIT and Politecnico di Bari may have just found a better way.

They developed what they call electrofluidic fiber muscles, tiny actuators about as thick as a toothpick that contract when electricity is applied, no motors, no external pumps, no noise. The whole thing works by injecting charge into a sealed dielectric fluid, which creates ions that move the fluid and generate pressure. The result is a fiber that behaves remarkably like real muscle.

What makes this genuinely exciting for robotics is not just that it contracts, it is that it can be distributed throughout a structure the same way muscle is in a body. Power density sits at around 50 watts per kilogram, on par with skeletal muscle, with a contraction response of 0.3 seconds. In tests, a bundle of these fibers lifted 4 kilograms, a woven version bent a robot arm 40 degrees, and a fast configuration launched objects in under a third of a second.

If this scales, the path to robots with real fine motor control gets a lot shorter. The bottleneck has always been the hardware. This starts to remove it.

why it matters

Most robots are built around servo motors that convert rotational force into linear motion and concentrate bulk near the joints. These fibers contract like real muscle and can be distributed throughout a structure. For exoskeletons and prosthetics, it’s the difference between gear you strap on and gear that moves with you.

Link to the article

Humanoid robots going mainstream is one of those things that feels like it is always five years away. But reading this, I got genuinely excited. Tesla thinks it has found its shortcut and the answer is Shanghai.

Allan Wang Hao, Tesla China’s president, called the Gigafactory a “golden key” for scaling Optimus production. And when you look at the numbers it is hard to argue. The Shanghai plant pushed out 851,000 electric vehicles in 2025, more than half of everything Tesla made globally. The supply chains, the assembly lines, the engineering muscle, it is all already there. Tesla wants to point that same machine at a humanoid robot with 40 degrees of freedom and see what happens.

The target price is somewhere between $20,000 and $30,000, aimed at home assistance and elderly care. If they hit that number at volume, this stops being a tech demo and starts being a real product category. Yes, Tesla has a long history of ambitious timelines that stretch well beyond the original promise. But the underlying idea, that automotive scale manufacturing could do for robots what it did for EVs, is genuinely compelling.

Shanghai has already proven it can build things at a pace and cost that most of the world cannot match. The question is whether a robot is close enough to a car for that to matter.

why it matters

Shanghai already delivers more than half of Tesla’s global cars, but whether or not Tesla could smoothly put Optimus into those same high-throughput lines isn’t quite clear. Still, Shanghai’s scale is its best shot at jumping from a few hundred prototypes a year to the tens of thousands it needs to make this business real.

Link to the article