If by spoofing you mean holding up a stop sign, I don't see this as a problem. Google's cars use Bayesian reasoning. There are many additional pieces of information for the car to draw on that should outweigh the evidence of the stop sign:
* a Google car will have been on that road before (many times) and never seen a stop sign (and Google can probably tap into the government data for new stop sign installments anyway)
* looking at a map the car can establish what the prior probability of a stop signing being there is (probably low if it's not an intersection)
* the car can look at cars in front of it and observe what they are doing
Beyond this a simple look ahead will solve the problem: the car can extrapolate what will happen if it does or does not stop. If stopping suddenly at the stop sign means causing a rear end accident, but not stopping at the sign has no negative effects (because there is no cross traffic), then the car can safely ignore the sign, especially if it has assigned a non-negligible chance to the sign being fake.
Google's cars use Bayesian reasoning. — How does that change anything? What's an alternative to "bayesian reasoning?"
Google can probably tap into the government data for new stop sign installments anyway — such a thing (an accurate government database of new stop signs) does not exist.
the prior probability of a stop signing being there — scenario: a construction worker with a rotating STOP/SLOW hand sign during road work. Or, manual traffic overrides with police standing in the road with hands up or hands waving to manually control traffic.
the car can look at cars in front — assumes said cars exist and the motion of forwardness is more important than any potential cross traffic being stopped for.
I think he meant that instead of following deterministic rule (ie stop sign == stop), the car can take into account other factors, and make the actions with the best probability to not cause a disaster.
And keep in mind that the world in which self-driving cars are dominant is very and much different than the world we're living right now (as far as transportation are concern). Stop signs are meant to be read for human, car don't actually have to care about it. How about instead of stop sign at crossroad, we have some devices that can monitor the incoming traffic, and relay that information to the car? (I'd say that the car can even communicate with each other by themselves - but that approach has a host of other problems too).
Honestly, I'd say the main problem in a world of self-driving car are other human-driven cars: the self-driving car has no idea what the human would do, and the human has no idea what the automated car would do.
RISK: Since each AV in the fleet represents an access point into such systems, it may be infeasible to create a system that is completely secure.
MITIGANT: "Fortunately, robust defenses should make attacks even more difficult to stage. The <U.S.> has demonstrated that it is possi ble to maintain and secure large, critical, national infrastructure systems, including power grids and air traffic controlsystems."
This is a combination of an assumption and non-sequiter.
* a Google car will have been on that road before (many times) and never seen a stop sign (and Google can probably tap into the government data for new stop sign installments anyway)
* looking at a map the car can establish what the prior probability of a stop signing being there is (probably low if it's not an intersection)
* the car can look at cars in front of it and observe what they are doing
Beyond this a simple look ahead will solve the problem: the car can extrapolate what will happen if it does or does not stop. If stopping suddenly at the stop sign means causing a rear end accident, but not stopping at the sign has no negative effects (because there is no cross traffic), then the car can safely ignore the sign, especially if it has assigned a non-negligible chance to the sign being fake.