Wouldn't it just burn more calories?

Isn't it just exercise? Unclear what the concern would be despite in general agreeing with you on human body being hard to change without side effects

Did you read even the first paragraph of the link? This is about using skin as a medium of power distribution to devices, not harvesting power from the human body. No hand-wringing required.

Makes sense...basically the energy equivalent of Herbert's https://dune.fandom.com/wiki/Stillsuit

Solar and body heat are overwhelmingly the most potent out of all of those. And even then the only reason I'm listing body heat is because of how often things will be in pockets or bags and therefore not exposed to sunlight — small heat differences are fundamentally inefficient to extract useful work from.

For motion: The "standard" calorific demands of a human work out as about 100-120 watts or so; adding some mechanism that extracts energy from your motion, makes your motion harder by that plus whatever gets lost as heat (which can't be extracted efficiently). Something that makes you burn an extra 500 kcal/day would at best be 24 watts, but whatever it was it would basically have to put continuous resistance on some of your movement to get that, like an exoskeleton but it doesn't assist you it just slows you down the whole time.

Now, someone correct me if I'm wrong, but I think human biological efficiency is around 25%? So that 500 kcal/day -> 24 watts actually looks like (6 watts of mechanical output + 18 watts of heat)?

And those 18… the theoretical maximum efficiency depends on the difference between the hot side and the cold side, so to be as efficient as possible you'd need to be somewhere cold. The Carnot efficiency limit is η = 1 - T_cold/T_hot (absolute temperature, i.e. in Kelvin), so human body temperature of 310 K (37°C), somewhere really cold 253 K (-20°C) -> 1 - 253/310 = 0.184 (18%), which gets you 3.3 W from an extremely unpleasant experience.

Even the full 24 W is about what you'd get from a T-shirt made from the best solar cells with a reasonable assumption about capacity factor.

The only places I'm aware of where wind beats muscle, is inside a hurricane or a tornado. And that's if you could harvest it on a size scale comparable to your body.

> Garmin already has solar on many watches to extend battery and the Kinefox is already doing kinetic on animal tracking

Some of the mechanical watches when I was a kid were advertised as keeping themselves wound from body motion. Apparently those predate my birth by over two centuries.

My school calculator back in 1995 was already solar powered, too.

Neither of these uses are power-hungry.

It's obviously useful right? Currently I have to remove my smartwatch to charge it for a tiny example

The paper answers this. It's about powering a suite of sensors and devices that can also communicate using your skin as the distribution system for power and comms. I could see the appeal in a world where many such sensors and devices might exist on a single body.