This is really easy to verify.
You drop an iPhone, the gravity "acceleration" really goes to 0 - so far so good.
But at rest (i.e. holding the iPhone in your hand), the acceleration is pointing downwards, not upward as she claims.
Try this thought experiment. Assume your phone is really accelerating downwards when at rest. Then let it enter a free fall i.e. drop it. Now it's accelerating down even more so the downwards acceleration should show an increase. Actually, it goes to zero. Relative to a free-fall, being stationary is an acceleration upwards.
The relative direction of the acceleration would be "up". This is what she means.
this is obviously a very confusing topic, as evidenced by the other replies that either don't get my post, or the post above :D
I think there's some sign inversion happening there. But don't quote me, all the details are fuzzy to me.
The example you give also doesn't make sense. The acceleration would not go to 0 when you drop something. The acceleration continues for the entire fall. It may be offset by friction if it hits terminal velocity. But, absent that, acceleration would continue the entire fall.
If the reading of an accelerometer does change in freefall, then what it is measuring isn't the acceleration, necessarily, but the difference in acceleration between components in the device. Which makes sense. Is like watching a balloon in a car when you start moving. (With the trick of whether the windows are open or not, of course.)
This would come as a surprise to the designers. :)
No, the acceleration on an object in free fall is in fact zero. It may be surprising, and may require rewriting your intuitions to fully grok, but this is indeed what GR says.
Edit: (I'm actually curious what I said is fully wrong, btw? Accelerometers measure how much an inner piece drifts to other things based on differential in acceleration, no?)
Edit2: I think you thought I was saying the accelerometer would not read zero in free fall? I meant more that you would continue accelerating, despite the reading being whatever it was.
In a steady state, all the components are accelerating equally. It measures the force between the components and uses that to compute acceleration, yeah, but the positional drift between the components is minuscule and only happens during jerk.
(This describes one type of accelerometer. I'm not up to date on every possible or currently used design.)
My point is that to an earth bound observer, something in free fall to the surface of the earth is accelerating faster to earth until they hit terminal velocity. This is despite the accelerometer on the falling body showing zero.
Note that i don't think this adds any understanding to whether or not gravity is a force. But it does feel a lot like saying that, if you are in a river and able to stay still, that you aren't fighting the force of the river. Similarly, do we say that people standing on a moving sidewalk are stationary in the same way as someone standing on the sidewalk they are passing?
That is extremely easy to understand but somehow you and all of these "eminent scientists" fail to grasp it.