Unfortunately, the data isn't consistent. That's why they need to make corrections. The question is "Do the corrections make it more nearly accurate?", and that's really hard to demonstrate. When there's too much noise in the signal, it's really difficult to filter it out without losing the signal.

When you're talking petabytes and "reading DNA", I don't think 60 minutes is the right order of magnitude.

I'm not sure this is going to be public facing. Many of the objections seem to assume that it is. OTOH, AI is known for returning the answers you want it to return, regardless of the truth of those answers. So perhaps it's the perfect yes-man.

I'm not going to claim those are harmless, but banning them appears more socially destructive than allowing them.

Going to a particular date on a tape is a seek operation. A better reply would be that there's more than on kind of cassette. (1/2" tape has been in cassettes before, just not the kind you usually think of. And that was durable enough to allow a reasonable number of seeks. But I'd sure hate to have to patch a tape with that density.)

But how long does it take to search to the particular bits you were looking for? How many times can you search through the tape before it breaks? There are reasons random access cassettes were never popular.

It's not hard to believe. There are lots of plausible explanations. Most likely some of the chips are multi-function,but sold to do one particular job (so only partially documented).

That doesn't imply that the chip didn't have a built-in radio, just that it couldn't transmit or receive without some ancillary mechanisms.
I suspect the claim is technically true, but the reason is that the chips were designed to be sold to different people to do different things.

Are you reasonably certain that this claim is false? Remember that the US is trying to do something similar to NVidia chips. Also lots of chips are multipurpose, with the same chip being sold to different people for different functions.

I'm not sure I believe it, but I'm also not certain it's wrong.

Yes, it will be largely human interaction, and action in environments designed for humans. But that's NOT a small use case.

Last I heard, the mouse models of Alzheimer's were really bad models.

The argument about usuform robots vs. humanoid robots goes back to at least the 1940's. We've *got* usuform robots, Humanoid robots are still works in progress (if you don't count things like the robot receptionist that built into the desk...mix that with a vocal chatbot and there are some jobs it could do. If I read right the one on display at the latest show in China didn't need to be built into the desk, and had a full humanoid body...attractive female, but clearly artificial, unlike the earlier one which wasn't clearly artificial, until you noticed why she couldn't stand up.)

IIRC, the first robot to take care of elderly (by making them feel better) was a cute little white furred seal pup. Dogs and cats can do it too. So that's not an argument for a humanoid robot. It *is* an argument for a robot that at least *looks* cuddly.

OTOH, would a robot dog be trusted if it tried to give the medicine? No hands, so all it could do would be spit it out.

For general purpose use, I think the humanoid factor is better, but probably best to stay on the far side of uncanny valley. Two specialized forms seen optimal, one aiming at "cute" and one aiming at "strong". The "strong" one would be for things like helping a patient out of bed, etc. The cute one would be for encouraging obedience..."It's time for your medicine." I suppose the cute one wouldn't need to be humanoid, but it *would* make things easier. And the "strong" one needs to avoid seeming threatening.

The reason such things haven't evolved is that our many-agos ancestors evolved from fish with four fins into tetrapods. Adding another pair of limbs is not something that's easy to do. It's much easier to lose the legs, like whales did.

How many humans can "sculpt like Michelangelo or paint like Rembrandt"?