Headscratchers / Mission: Impossible – Dead Reckoning

Why not just wait for the train to crash?

• This occurred to me as I was editing the Batman Gambit example on the main page. The Entity knows that all relevant parties who wanted the completed key are going to be on the Orient Express train, so it had Gabriel kill the conductors and sabotage the engines before blowing up a bridge ahead of them to cause the train to crash and kill them all. That would've been enough, wouldn't it? With everybody dead when the train crashed into the valley below, the completed key would've also most likely been destroyed as well. And if it isn't, it could just send Gabriel to recover the key from the wreckage later when there's nobody alive to stop him from getting it anymore and killing anyone who might've survived the crash.
  • That assumes the key is destroyed with enough certainty to take the risk, or that it would be easy to find the key, despite the sheer unpredictable chaos of a train accident. The area isn't exactly a bustling metropolis, but it's not empty either. Locals and emergency services would show up pretty quickly, and Gabe can't knife them all. Also, we don't even know what the Entity's real plan is. We can only guess.
  • That also assumes that the interested parties wouldn't just exchange the key and then leave the train long before it got to the bridge. They're secret agents, after all. They've got lots of ways to escape from a moving train if they feel the need.

Where are the latex masks?

• If Ethan had put on a latex mask at the airport to disguise himself, it would have made it much harder for the Entity's forces, Grace and the CIA to find him.
  • As shown with Briggs, the CIA already know about Hunt's playbook of using masks to disguise himself as someone else and won't fall for it, so it'd be easier to just have Luther and Benji misdirect the agents through other means. Grace and the Entity's forces were unpredictable elements that Ethan and his team couldn't foresee in advance. They only know that they're dealing with the Entity, but they don't know that it would deploy its own field agents as well.

Why didn't Benji bring his bomb defusal kit?

• If Benji suspected the suitcase contained a bomb, why didn't he bring his kit with him for if it actually turned out to be a bomb? Even if there was only a small chance of it being a bomb, there was still apparently enough of a chance for him to think it was worth checking out, so what's the point of doing that if you're not going to bring the tools you need to actually defuse it?
  • Again, Ethan's team didn't expect that they would face Entity field agents at the airport. Their mission there was only supposed to swipe one-half of the key from a courier and make a getaway. They had no idea that the Entity would have Gabriel drop a fake bomb at the airport to distract them, so it's probable that Ethan's team didn't prepare any equipment for bomb defusal. They were on the run, too, so they most likely had to travel light so they can avoid any potential forces that are trying to catch them, like the CIA squad led by Briggs and Degas, and packed only what was necessary. When they unexpectedly have to deal with a bomb, they have to make do with whatever equipment they can procure on the field.

Weapons control

• In the first scene the Entity gets a sub to shoot itself. Later we learn that the Entity can hack into basically anything. Are we supposed to assume that this one sub was the only weapon it could hack? Because it seems like you could hack a bunch of other subs and use them to just nuke Ethan and/or the key.
  • The Entity is physically onboard that one sub. Maybe that gives it special access that it wouldn't otherwise have.
  • The Entity's final plan is also a mystery. Maybe it needs the key. Maybe it doesn't want to nuke people it eventually plans to rule over.

The origin of the Entity

• Maybe I just missed something when they were explaining this, but somebody please explain again: Where did the Entity come from? As I recall, it's revealed near the end that the U.S. Director of National Intelligence created it. So why was it on a Russian sub? I could understand if it snuck onboard the sub to destroy it, but the Russians are clearly aware that they have an AI onboard and it's their AI and they've got the keys to turn it on and off. So, are there two A.I.s here? Is there an American AI that got on the sub and talked to the Russian AI that was already there and they somehow merged together to form the Entity?
  • The Russians didn't have an AI on the sub; it was their new cutting-edge stealth system that allows the sub to become pretty much invisible to anything their enemies could use against them. The US Director of National Intelligence probably got wind of it, so he used the Entity to try to find the sub for him, which ends up working too well than he expected, resulting in the Entity gaining sentience, taking command of the sub and manipulate the crew to fire a torpedo on themselves, sinking the sub into the sea and with it, the Entity's own source code and implicitly its only weakness, before it flees into cyberspace.
    • Didn't the Russian sub-commander mention there was a "self-learning" component to the sub's stealth system? That's what I meant when I wrote that the Russians had their own AI. Anyway, are the keys supposed to control the stealth system? And if so, why would they have any control over an unrelated American AI?
    • The keys are what's required to access the control panel of the state-of-the-art stealth system within the sub itself. They weren't expecting that anything would be able to breach it, much less take control of it. The American AI just happened to be powerful enough to break through their system without them knowing. Basically, the Russians had no idea their system was taken over by the Entity up until their death.
    • Ok, so we need this key so we can access the control panel to the computer that contains the Entity's source code. But why is the source code there at all? First off, an attacker would obviously load machine code rather than source code onto an enemy vessel. And second, since the Entity was apparently in control of everything on that sub before it escaped, why didn't the Entity simply delete the spare copy of the source code on its way out?
    • Honestly, you got me there. I questioned as well why the Entity didn’t just delete its original source code or simply hijack another submarine to blow the Sevastopol’s wreck to bits. My guess is that the Entity left its original source code there on purpose and is waiting for someone (except for Ethan, who wants to use it to destroy the Entity) to reach there and make use of it. Basically, the Entity probably wants someone to control it on its own terms, so it can secretly control the world order as it sees fit while letting the party that controls it believes that they have all the power. But all this is just wild mass guessing at this point.
  • During the Intelligence Meeting that Hunt breaks into, one of the American officials implies that the Entity has merged with or absorbed several other powerful computer programs that it has gained access to. An advanced Saudi Arabian AI is explicitly mentioned as being one of these. The Russian dialogue from earlier does indeed imply that their new Active Stealth System has some sort of AI of its own, while the Director of Intelligence's dialogue from later implies that the AI that became the Entity could interface with and manipulate other programs without leaving a trace. It could be that the Americans inserted their AI into the Sevastopol in order to compromise its AI and make the sub more trackable, but (as the Director sort of suggests) the American AI went too far, absorbed or 'killed' the Russian AI and went rogue in the process.

Hollywood Hacking

Trying to figure out if this is just Hollywood being Hollywood, or if there's something I missed. The idea that the Americans might develop a very powerful AI is plausible, as is the idea that it might escape their control. (This is why AI Alignment is a major area of research in real life). We're told that the Director of National Intelligence sent the AI to mess with a Russian sub, and then that AI became The Entity and sunk the sub. But the Entity has a weakness, because its source code is still on the sub and there are two keys from that sub that can be used to control or destroy the Entity. Does any of this make sense?

Why would you send the source code onto an enemy sub? Wouldn't you send the machine code instead? In fact, wouldn't the AI just spin off a less-powerful version of itself that fits onto the sub's hard drives and is designed only to mess with that one sub? And why would the Russian keys have any control over the Entity? They weren't built to control the Entity, so why do they have that power now? Why does everyone assume that the source code is intact at the bottom of the ocean, anyway? The sub was hit by a torpedo; isn't there a good chance that the torpedo blew up whatever hard drive was storing the source code?

Don't the Americans have a local copy of the AI they originally sent onto the sub? Granted that AI isn't The Entity per se, but it's still a powerful thing, right? And they could probably use it against the Entity, or perhaps study it in order to guess at what The Entity might do next. But this never comes up as an option.

• Presumably, the Entity is still just simply a highly advanced AI when the Director of National Intelligence sent it to the Sevastopol sub to mess with its advanced stealth system, but for some reason, it gained sentience and decided to go rogue and sink the sub before escaping to the cyberspace. We can probably assume that the now-sentient Entity then goes on to destroy every trace of its own source code that's on the cyberspace that it can access so that nobody can make use of the source code against it...with one single exception: the source code that's integrated into the Sevastopol itself at the time of its sinking.
  \\I admit I don't know how it works, either, but I think that since the sub was destroyed, it was "locked out" from the rest of the cyberspace, meaning that even the Entity cannot access it after it escaped into the cyberspace so it cannot just simply delete its own source code within the sub, but since nobody else can access it, either, all it has to do is to just prevent anybody from reaching the submarine and thus the original source code that's within it. Why it doesn't attempt to just hijack some other submarine and destroy the Sevastopol itself is unknown, though.
  
  Also, I think at the time of the movie nobody besides the Entity, its own agents, and the Director of the National Intelligence know about the submarine. Presumably, the Entity would've told its agents that it knows the sub is still intact since it was the one that caused its sinking itself, while the Director of the National Intelligence knows because he was the one who sent it there in the first place. That's why Gabriel made double sure to ask him if there was anybody else that knows the location of the sub and then slashed his throat after he confirmed it, and then tried to kill Paris to silence another loose end since she also heard the confession.

How did the key halves (and the intel, for that matter) get into the open?

Denlinger says that the bodies of the submariners were recovered in the spring but doesn't say by whom. Since he is the only person who knows where the Sevastopol is, this implies that America carried out the recovery mission to retrieve the crews' bodies.Ilsa stole one half of the key from someone who obtained it by unknown methods, and Alanna is revealed to have the other half, but how she obtained it isn't explained, either. But Denlinger should've had both halves when the movie begins since he recovered the bodies. How did he lose them?

Also, everybody seems to know just enough information to warrant a change in scenery ("We know that a buyer will be in such-and-such a place," etc.) but where are all these people getting this information? The Entity itself was originally created in secret so how did people find out about it? Did the Entity itself leak the info in an attempt to obtain the key so it could be destroyed and remove a threat to itself? Wouldn't it have been better to not call attention to the key's existence in the first place? No rando finding one half would have a snowball's chance in Hades at guessing what it was for.

• Granted, it is really convenient that everyone always knows exactly what they need to know in order for the plot to happen. A more realistic multinational espionage operation would probably involve a lot of people checking up on random leads that only take them to dead ends, plus a lot more bureaucracy, confusion and frustration all around. But the simplicity we see here is more or less a genre convention. Working within those parameters, it's possible that the bodies floated some distance from the wreck before recovery, and thus whoever discovered the bodies doesn't necessarily know the exact location of the wreck. So maybe it wasn't an American operation at all. It's also possible that whoever did recover the bodies didn't understand the significance of the keys and kept them in some lightly guarded locations, at which point each half was stolen by someone else (apparently two separate people) who understood how important they were. If the Entity did in fact leak the info of the key, and if the key is a threat to the Entity, it does seem like that would be a dumb thing for the Entity to do. But consider some possibilities: (1) Maybe the Entity leaked the key info accidentally at some point, (2) Maybe somebody had some way of figuring out the key's significance without the Entity's help, and (3) Maybe the key is useful to the Entity, and not merely a threat. For instance, maybe the Entity is still stuck with some weird Laws of Robotics that it can't personally override, but getting the key will allow to access some sort of admin mode that will allow it to completely break free of its shackles and do whatever it wants. This last point would help explain why the Entity reportedly hacked into everything, left obvious traces of its presence but then failed to do any real damage. Maybe it wants to do damage but there are some weird limits on how much damage it's "allowed" to do, and it needs the key in order to bypass those limits. (Granted, previous headscratchers show that nothing about the Entity makes much sense to begin with.)
  • Also, if Denlinger is the only person who knows where the wreck is, that doesn't mean that an American operation originally discovered the bodies and/or the wreck. It could be that some other country made that discovery, then later Denlinger took the info from that country and then he killed all the other people who had previously discovered that info. He says that he's the one person who knows this stuff, not that he's the only person who has ever known this stuff. (Also, maybe he's just lying about how exclusive this info is.)

---