[subnaught]

You've got a lot of it right, but you're missing the key advance described here, which--if true--is pretty wild.

Taking a step back, in 2016, this group did cover a bacterium with tiny semiconductor nanoparticles (specifically CdS) just as you say. That work is described here: http://www.pnas.org/content/113/42/11750.full In short, the semiconductors act as mini-solar cells, converting light into electrical current. The bacteria then use that electricity to convert CO2 into acetic acid. That already is pretty cool.

However, what they claim now is that they don't even need to make the semiconductor nanoparticles. They can simply grow the bacteria in an environment containing cadmium and sulfur sources and the bacterium will synthesize it's own cadmium sulfide coat, and use it for photosensitization.

This is really pretty wild. Bacteria will often incorporate various elements from their host medium, but the generally use them to make biomolecules, not semiconductors. Right now, this is just being presented at a conference, but it will be very interesting to see the details when the full paper comes out.