It's completely wacky, but if you've got a few mil to drop on wacky ideas, it starts to seem a bit less wacky in the grand scheme of things. It's basically a wager: "There's a small, but perhaps nonzero chance I'm living in a simulation. I've got the throwaway cash to fund an escape attempt. This investment has a massive chance of being completely wasted, even if this is a simulation. But if there is even a tiny chance of figuring something out, I might as well take it."
What would it even mean to 'escape' a simulation? In my mind I'm imagining when I write code to model a real process and even at some super high-fidelity simulation I can't imagine an 'escape'. Maybe if the simulation were so self-aware it could conduct local privilege-escalation attacks and replicate itself like a worm/virus?
EDIT: aside -- "Microcosmic God" [1] was a great short story along these lines
Assume a non omnipotent being, with non-infinite resources is creating a universe simulation. What decisions would we expect a person like this to make, when designing this simulation?
Since they do not have infinite resources, they would likely take shortcuts, and make compromises and tradeoff when designing the simulation
For example, if light did not have a speed limit, and interacted with the entire universe instantly, this would be very hard to simulate. You would save resources by giving light a speed limit.
We also expect that they might abstract out the nitty gritty details of how things work at the atomic level. If you don't really care that much about how very small particles interact with each other, you might have these numerous calculations evaluated lazily, when they are "observed" by the higher level entities.
Things get really interesting when you start thinking about what kind of bugs a god programmer might be likely to make when designing their simulation.
Are there any weird natural processes, that would be much more elegant if our equations modeling them were changed very slightly, almost as if the equation was a mistake in the first place? Another attack avenue is to combine two natural processes in unexpected ways, to try and find "edge cases".
How do you know your own thought processes aren't a part of the simulation? Logic is a great tool to analyze the simulation unless it's been baked on purpose by our overlords.
On a different note, does simulation hypothesis strike anyone else as the idea God just explained in pseudo scientific/computing terms?
The difference is that God is an unfalsifiable claim.
The simulation argument is one that can be supported or argued against using evidence.
For example, if scientists find some new nature process that would be very difficult to simulate, (infinite speed of light is one such example) then that is strong evidence that we aren't in a simultion.
Of course, even if we keep observing things about the universe that would make it easier to simulate, that doesn't mean we actually are. It could just all be a coincidence. Or maybe it doesn't make any sense for a universe to be difficult to simulate.
The basic idea is to ask the question "Assume someone lives in a simulation. What would that person be likely/unlikely to observe?". and "Assume someone does NOT live in a simulation. What would they be likely/unlikely to observe?" And see how much this stuff matches up with reality.
When you find an exploit that lets you execute any code you want on a remote server, you can find out all the internal details of that server, and you can make that server do whatever you want.
That's actually a pretty cool idea for a movie. Scientists hacking the universe to achieve alchemy and time travel. Maybe it could be a sequel to the Matrix. It's strange how they only ever made that one Matrix movie.
Essentially that's what Agent Smith does in the Matrix sequels. He can't "escape" the Matrix, but he can basically make it his own. (For the purposes of this analogy I'm intentionally ignoring the ridiculous scene where he somehow transfers himself to the physical world.)
Interesting to consider: if we ever discovered ourselves to be living in a simulation and attempted to do something about it, would the simulator simply reset the simulation (a la The Matrix trilogy)? Or would the simulator be intrigued by our self-awareness and allow the simulation to continue?
But anyhow, that's enough of a digression for one day.
'accelerando' had a great throwaway line about that:
And then there's the weirdness beyond M31: According to the more conservative cosmologists, an alien superpower – maybe a collective of Kardashev Type Three galaxy-spanning civilizations – is running a timing channel attack on the computational ultrastructure of space-time itself, trying to break through to whatever's underneath.
Really.. It doesn't make sense. This is like the Star Trek episode where a holodeck character wants to live in the real world. (Spoiler alert..) Picard just traps him in a bigger simulation and makes him think that he escaped!
