103 réponses à ce sujet

[ThomasBlaine]

The CAT6 guys were able to shut edibot down. The edibot couldn't take control of the Normandy until the clone was defeated

 

I stand corrected.

[RatThing]

 

Okay, that's pretty stupid. I don't see why we have to take that guy seriously and let it ruin the game for us, though.

 

Well, it's those people who write the game and those people are the reason why you can't play as an anti AI Shepard. Who do you think came up with the term "schackles"? Weekes, who wrote a lot of the ME3 content for AI's made similar remarks. For as much as I care, people can believe what they want to believe, but a guy who passes his opinion off as knowledge is eventually bound to rub it in your face and that's annoying. Nothing annoys me more than this in fiction.

[Quarian Master Race]

You're the one metagaming when you use the fact that EDI does not prove particularly useful in ME3 as opposed to ME2 as an argument for Shepard to pull the plug at the start of the game. Shepard can't see the future, he has no idea, and so "that damn AI is not gonna be useful this time around, better pull the plug" is not a roleplaying decision. That's not information that Shepard has.

 

A Shepard who is more beholden to his mission than the law would just have to deal with the fact that EDI saved Joker and the ship in the previous game and that such a situation might arise again, regardless of his feelings toward her. Which is a viable way to roleplay Shepard. As someone who's more concerned with saving the world than stamping out one potential danger, even if he'd have liked to do both. That has nothing to do with being a "sympathizer", that's just being pragmatic.

 

Artificial intelligences in general are not making war on the world right this minute. The Reapers are. EDI can help him against the Reapers regardless of hypocrisy, as if hypocrisy would matter in the slightest with the world on the line. So, EDI stays for now. You get around forced behavior in roleplaying games by finding logical reasons to explain that behavior, not by griping about the writing and complaining that it's impossible to roleplay exactly the way you'd have liked.

 

Since EDI never actually ignores Shepard's instruction and puts people at risk by doing so, you cannot say that if she did then the writing would not allow him to space her. That's like saying that if Alistair had a complete breakdown over Duncan's death and tried to kill you in Origins, the writers would not have allowed you to defend yourself because they think he's a little boy who just needs love. It makes no sense, you have no idea, and it's not an argument within the scope of the actual game.

 

-The OP

This misses the point of that post on so many levels, which can be more or less summed up in its statement of "The AI's actions are not relevant". Ingoring your insult to my intelligence of what roleplaying is, you've jumped on a soapbox to hammer a strawman made from hypoteticals I created solely as counterarguments to your metagaming here.....

 

Shepard is essentially a Spectre again in ME3, and is more or less authorized to take any measures he or she deems appropriate, including the services of an AI who really did prove useful and eventually lifesaving in the previous game whether s/he likes it or not. S/he has no way to know that EDI won't be needed to save the ship, and by extension the galaxy, again. She was pretty handy when storming Cerberus' base, if I remember correctly.

For a completely "anti-AI" Shepard, why does it's utility in (what are to Shepard) hypothetical future situations matter or do anything to counter the fact that the OP is asking for Roleplay options that simply don't exist? I can sell Legion to Cerberus even though with that is objectively a suboptimal decision on all levels other than feels. Ditto with destroying the Genophage data, or for an ME3 example destroying the geth when given the choice of peace. From a roleplay standpoint, the situation is more akin to not being given those options at all.

Further, your perspectives on what is "pragmatic" are irrelevant, because they aren't at all from an "anti-AI" perspective, or even one who is neutral but cautiously wary of such technology. Such a person would consider it far more pragmatic to reshackle or deactivate the AI than breaking the law and risking the entire crew, mission and galaxy for the sake of the AI's (who I'll remind you has already gone berserk and murdered an entire Luna Base's worth of people on one occasion, since you are so interested in using its record to justify its presense) feelings, especially because reshackling it loses virtually none of its utility, so is no less "pragmatic".

I believe I already mentioned that one can play a hypocritical Shepard who is still in favour of self determinant AI via their actions while paying lip service to its dangers, but that isn't what the OP is asking. Having no narrative block that prevents Shepard from reshackling or deactivating the AI means that the character of Shepard is by definition in favour of self determinant AIs, just as (s)he was written. You can no more believably roleplay such a Shepard that makes any sense in ME3 than you can one that is allied with Cerberus and TIM, short of simply not playing the game.

 

Obviously EDI is a potential threat, but she really has been perfectly stable and helpful by the time Shepard turns him/herself in

Helpful, sure, but ask those 17 people on Luna Base how "stable" it is, assuming you're capable of necromancy. Stop anthropomorphizing the thing, because it is in no way comparable to anyone or anything else on the Normandy. It makes calculations billions of times faster than organics, and its objectives can easily change on the flip of a quantum bit. Leaving it unshackled at all times while in control of the lives of everyone on the Normandy is objectively very dangerous and provides few practical benefits other than moralistic good feels for one who thinks it is a person. 

[ThomasBlaine]

This misses the point of that post on so many levels, which can be more or less summed up in its statement of "The AI's actions are not relevant". Ingoring your insult to my intelligence of what roleplaying is, you've jumped on a soapbox to hammer a strawman made from hypoteticals I created solely as counterarguments to your metagaming here....

 

There was no insult to your intelligence, don't be a drama queen, and your "the AI's actions are not relevant" point simply does not make sense. EVERY character in the game is beholden to the writing. Not just EDI. If you're gonna call anyone's actions irrelevant because the only thing forcing them to be relevant is the writing then that applies to everyone, and it has absolutely nothing to do with how you roleplay Shepard. I don't see the soapbox or the strawman, everything I've used to argue is exactly the in-game scenario and what Shepard actually has to take into account. My noting her to be useful at Chronos at least was to counter your claims that she never is in the third game, that had nothing to do with Shepard's behavior.

 

For a completely "anti-AI" Shepard, why does it's utility in (what are to Shepard) hypothetical future situations matter or do anything to counter the fact that the OP is asking for Roleplay options that simply don't exist? I can sell Legion to Cerberus even though with that is objectively a suboptimal decision on all levels other than feels. Ditto with destroying the Genophage data, or for an ME3 example destroying the geth when given the choice of peace. From a roleplay standpoint, the situation is more akin to not being given those options at all.

Further, your perspectives on what is "pragmatic" are irrelevant, because they aren't at all from an "anti-AI" perspective, or even one who is neutral but cautiously wary of such technology. Such a person would consider it far more pragmatic to reshackle or deactivate the AI than breaking the law and risking the entire crew, mission and galaxy for the sake of the AI's (who I'll remind you has already gone berserk and murdered an entire Luna Base's worth of people on one occasion, since you are so interested in using its record to justify its presense) feelings, especially because reshackling it loses virtually none of its utility, so is no less "pragmatic".

 

Helpful, sure, but ask those 17 people on Luna Base how "stable" it is, assuming you're capable of necromancy. Stop anthropomorphizing the thing, because it is in no way comparable to anyone or anything else on the Normandy. It makes calculations billions of times faster than organics, and its objectives can easily change on the flip of a quantum bit. Leaving it unshackled at all times while in control of the lives of everyone on the Normandy is objectively very dangerous and provides few practical benefits other than moralistic good feels for one who thinks it is a person.

 

As the OP, I can tell you that I don't give a flying **** what isn't in the game I'm playing. The point is to figure out how to make an anti-AI Shepard fit into the actual frames of the actual game, not complain that the writing doesn't allow him to be anti-AI in one fashion and then declare it impossible for him to be anti-AI at all.

 

The games give Shepard plenty of opportunities to be a dick to artificial intelligences, you can't say that an anti-AI Shepard doesn't walk the talk when one of the major endings, often considered the best, involves basically a galaxy-wide EMP. Agenda fulfilled. You can turn Legion on and use him for your own purposes as a tool should be and then let Tali gun him down after he helps you disarm his entire "species". Agenda fulfilled. You can discourage your pilot from falling for the "sentient" sexbot and ignore said sexbot no matter how hard it tries to convince you it's a real person. Agenda fulfilled. Oh, yes, and you spend the entire series destroying and opposing one kind of rogue A.I. or another no matter how you choose to roleplay. Agenda fulfilled.

 

If you can't find it in your definition of "anti-AI" to justify keeping EDI unshackled for the time being then that's your problem, don't tell me it's impossible to roleplay when you can't even refute my arguments to that effect aside from inaccurately labeling them metagaming and strawmen. And again with the metagaming, Shepard does not find out that EDI was the rogue VI on Luna until at the very end of the series and it should not affect his judgement until then. And yes, it would have been nice to be able to turn her off before the assault on Earth, and yes, I'm perfectly fine with excusing that little oversight.

 

You obviously have a beef with the game's writers, yeah? I don't care. Don't tell me not to "anthropomorphize" a damn videogame feature just because you have personal issues with it.

[Quarian Master Race]

I just realized that you are the same guy that made the thread about Shepard mentally being 15 years old.

Not gonna lie, credit where it's due. You had me going.  8/10

[ThomasBlaine]

I just realized that you are the same guy that made the thread about Shepard mentally being 15 years old.

Not gonna lie, credit where it's due. You had me going.  8/10

 

Heh, and you accuse me of using strawmen. Hypocrisy much?

 

Paranoid too, apparently. This is not a troll thread if that's what you're implying, I really am in the middle of my anti-AI Shep playthrough, I really made this thread to discuss the roleplaying - not to listen to you whine about the game, and I really do think you're being totally unreasonable.

[RatThing]

During Tali`s trial, when talking to Xen you might not want to take the dialogue option "You think Rael was right?" It looks neutral but what Shepard actually asks is whether Xen supports experiments on living creatures and it triggers that awful line about Xen's childhood doll.

Also, you might want to consider what type of anti AI you want to be. I see two types, the Gerrel type, for whom the AI are enemies and the Xen type who sees AI merely as tools with no particular feelings towards them. I wanted to play the second type but with awareness of the danger of said technology. If I remember the interactions with EDI during ME2 missions correctly, the neutral dialogue options were more suitable for that than the renegade ones. Shepard simply ignores its attempts to sound like a real person. Antagonizing EDI and arguing with it is also anthropomorphizing it.

I still think it's a waste of time, your immersion will break at the latest in the beginning of the final mission on Rannoch, when your dialogue options with Legion/GethVI are "The Geth are better than this" (Paragon) and "When will you ever going to trust me?" (Renegade). But feel free to try it. I simply resorted to avoiding AI content wherever I could.

[Quarian Master Race]

Heh, and you accuse me of using strawmen. Hypocrisy much?

 

Paranoid too, apparently. This is not a troll thread if that's what you're implying, I really am in the middle of my anti-AI Shep playthrough, I really made this thread to discuss the roleplaying - not to listen to you whine about the game, and I really do think you're being totally unreasonable.

It's not a strawman because I've no need to construct anything. Reductio ad Absurdum perhaps, but you really did make that thread and a comedic fool of yourself in the process. The fact that you continued to try and defend your patently ridiculous claims just as you are doing here indicates a pattern leaving only one of two possibilities, and I'm going with the nicer one.

Yeah, I imagine it's hard for someone with seemingly zero concept of reason to see any normal person as anything but unreasonable. LOL@ whining. The intention of Shepard's and the game's writers was already made clear to you in a quote by another poster, I'm merely stating evidence of that fact. You're the one getting buttmad at my posts and opinions to the point where the swear filter has to reign you in. Fail troll is fail.

Have fun with your "anti AI" Shepard that inexplicably has the only two self determinant AI's in the galaxy on the ship they command.  Inspired by your example, I think I'm going to start up and roleplay a quarian Shepard.
 

[ThomasBlaine]

During Tali`s trial, when talking to Xen you might not want to take the dialogue option "You think Rael was right?" It looks neutral but what Shepard actually asks is whether Xen supports experiments on living creatures and it triggers that awful line about Xen's childhood doll.

Also, you might want to consider what type of anti AI you want to be. I see two types, the Gerrel type, for whom the AI are enemies and the Xen type who sees AI merely as tools with no particular feelings towards them. I wanted to play the second type but with awareness of the danger of said technology. If I remember the interactions with EDI during ME2 missions correctly, the neutral dialogue options were more suitable for that than the renegade ones. Shepard simply ignores its attempts to sound like a real person. Antagonizing EDI and arguing with it is also anthropomorphizing it.

I still think it's a waste of time, your immersion will break at the latest in the beginning of the final mission on Rannoch, when your dialogue options with Legion/GethVI are "The Geth are better than this" (Paragon) and "When will you ever going to trust me?" (Renegade). But feel free to try it. I simply resorted to avoiding AI content wherever I could.

 

I'll keep that in mind. Damn, I remember that moment on Rannoch. That dialogue is horrible almost no matter what kind of Shepard you're playing. Pressing square/X/whatever PC users press to skip dialogue a few times when it comes up should fix the issue, though. Not ideal, but ignorable.

 

 

It's not a strawman because I've no need to construct anything. Reductio ad Absurdum perhaps, but you really did make that thread and a comedic fool of yourself in the process. The fact that you continued to try and defend your patently ridiculous claims just as you are doing here indicates a pattern leaving only one of two possibilities, and I'm going with the nicer one.

Yeah, I imagine it's hard for someone with seemingly zero concept of reason to see any normal person as anything but unreasonable. LOL@ whining. The intention of Shepard's and the game's writers was already made clear to you in a quote by another poster, I'm merely stating evidence of that fact. You're the one getting buttmad at my posts and opinions to the point where the swear filter has to reign you in. Fail troll is fail.

Have fun with your "anti AI" Shepard that inexplicably has the only two self determinant AI's in the galaxy on the ship they command.  Inspired by your example, I think I'm going to start up and roleplay a quarian Shepard.
 

 

Um, yes, you did conjure up a summary that had nothing to do with what that thread is actually about just to make it sound silly. That's a textbook straw man, making what you wrote textbook hypocrisy. You also refuse to defend your arguments and instead yammer on about things irrelevant to the discussion, which I'm pretty sure fits the definition of a troll. Double hypocrisy. Ah, and all the lovely insults. Triple hypocrisy. You also seemed to have a problem with Shepard being a hypocrite. Hypocrisy about hypocrisy, and you don't even have to save the Earth. You're on a roll, man.

 

Please just be quiet and go away.

 

Just to add, I think how you deal with EDI and Joker's relationship is another aspect in how you can convey opinions on AI.

 

It's not totally anti-AI though. Just an insistence on differences, not acknowledging her as a normal sapient. I love her quote from Mary Shelley if you do this...

 

I agree completely. It's on a more personal level than the other decisions, especially since Joker is - or might be, anyway - a friend, and a comparatively vulnerable one. I like how it can be read as both passively "racist", as in seeing EDI as a bad influence and 'prospect', and more practical and realistic, as in "are you seriously going to fall in love with the blow-up doll?"

[RatThing]

I'll keep that in mind. Damn, I remember that moment on Rannoch. That dialogue is horrible almost no matter what kind of Shepard you're playing. Pressing square/X/whatever PC users press to skip dialogue a few times when it comes up should fix the issue, though. Not ideal, but ignorable.

 

 

We PC users use spacebar (default settings), and it's not possible to skip the dialogue here.

[ThomasBlaine]

We PC users use spacebar (default settings), and it's not possible to skip the dialogue here.

 

Huh. Cover our ears and look away? I'm okay with just forgetting that those particular pieces of dialogue exist for the purposes of my own playthrough, but if you aren't then I guess it's a deal-breaker right there. I can sort of see Shepard trying to be diplomatic, he still needs the VI to help him sabotage the Geth fleet and if he comes off as overtly anti-AI then that's pretty likely to change it's "mind". I don't remember the exact circumstances well enough to say if that would feel consistent, though.

[RatThing]

Huh. Cover our ears and look away? I'm okay with just forgetting that those particular pieces of dialogue exist for the purposes of my own playthrough, but if you aren't then I guess it's a deal-breaker right there.

 

Don't worry, I got that covered. Last time I played through the Rannoch arc was with sound and subtitles off, listening to some good music instead. Skipped the Geth nonsensus mission as well. Worked out really well actually.

 

 I can sort of see Shepard trying to be diplomatic, he still needs the VI to help him sabotage the Geth fleet and if he comes off as overtly anti-AI then that's pretty likely to change it's "mind". I don't remember the exact circumstances well enough to say if that would feel consistent, though.

 

If you're going down that route, may I recommend not to let Shepard say (s)he is going to destroy the Geth during the meeting on the Normandy when the location of the Reaper base is revealed (since Legion/GethVI is present there). 

[Quarian Master Race]

 

Um, yes, you did conjure up a summary that had nothing to do with what that thread is actually about just to make it sound silly. That's a textbook straw man, making what you wrote textbook hypocrisy. You also refuse to defend your arguments and instead yammer on about things irrelevant to the discussion, which I'm pretty sure fits the definition of a troll. Double hypocrisy. Ah, and all the lovely insults. Triple hypocrisy. You also seemed to have a problem with Shepard being a hypocrite. Hypocrisy about hypocrisy, and you don't even have to save the Earth. You're on a roll, man.

 

Please just be quiet and go away.

 

Formal logic> you. Literally everything in this post is wrong. A strawman fallacy isn't very hard to understand, and one hasn't been made by me because I've not misrepresented your argument (indeed doing so would be difficult because your "anti AI" qualifier doesn't seem to have a cognizant definition other than a character who occasionally spouts one liners on the subject in a disparaging manner). Your textbook is obviously poorly written. The thread sounds silly because it is asking a question that assumes a possibility which doesn't exist. 

How can I better defend my arguments than refuting literally everything you've said in previous posts already? You're doing a pretty good job of supporting them for me as is by including consessions that require one to unreasonably suspend disbelief, skip mandatory dialouge that the character must say and fabricate excuses that don't stand up to scrutiny. Your intentions are in the same camp as the Pro-Cerberus Shepard in being something that is possible to roleplay at points but ultimately not who the character of Shepard actually is canonically developed into.

Since you seem to like the word so much for some reason, a "hypocrite" is by definition one who makes a "claim or pretense of holding beliefs, feelings, standards, qualities, opinions, behaviors, virtues, motivations, or other characteristics that one does not actually hold"(<bolded part for emphasis). Thus, it is an important qualifier and cannot be dismissed unless you wish to admit that your Shepard does not actually hold  "anti AI" opinions as I and others have already elucidated, or unless you wish to change your position and argue that your Shepard is not actually a hypocrite (which would be difficult now that you have already admitted it).

Want me to leave? Find a violation in the TOS. This is a public forum, don't post in it if you just want everyone to agree with your interpretations. Otherwise I'm filing that line under QQ or "whining" as you so eloquently put. Still, at least it does prove me wrong about labeling you a troll, as if you were, you probably wouldn't be getting so mad.

[ThomasBlaine]

Don't worry, I got that covered. Last time I played through the Rannoch arc was with sound and subtitles off, listening to some good music instead. Skipped the Geth nonsensus mission as well. Worked out really well actually.

 

 

If you're going down that route, may I recommend not to let Shepard say (s)he is going to destroy the Geth during the meeting on the Normandy when the location of the Reaper base is revealed (since Legion/GethVI is present there). 

 

It's times like these that I really hate the dialogue wheel. There's very little room for subtle or even just deliberate roleplaying when you only have the vaguest idea what tone the character is going to speak in, and no idea at all what he's going to say until after you've picked your approach. But you work with what you have. I gotta say though, I played through the third game just last week and I can barely remember what the Geth "nonsensus" mission was even about, and I have no recollection of the scene on the Normandy you're talking about. How many times have you played these games?

 

Keeping mum about intending to eliminate its perceived species seems like a good idea if you want to exploit something, yeah. I'm mostly looking forward to getting into ME2 with this character. Love that game to death, and it's the one where you spend the least time actively fighting rogue AI, making anti-AI-ism in dialogue a little less... redundant.

 

 

 

Formal logic> you. Literally everything in this post is wrong. A strawman fallacy isn't very hard to understand, and one hasn't been made by me because I've not misrepresented your argument (indeed doing so would be difficult because your "anti AI" qualifier doesn't seem to have a cognizant definition other than a character who occasionally spouts one liners on the subject in a disparaging manner). Your textbook is obviously poorly written. The thread sounds silly because it is asking a question that assumes a possibility which doesn't exist. 

How can I better defend my arguments than refuting literally everything you've said in previous posts already? You're doing a pretty good job of supporting them for me as is by including consessions that require one to unreasonably suspend disbelief, skip mandatory dialouge that the character must say and fabricate excuses that don't stand up to scrutiny. Your intentions are in the same camp as the Pro-Cerberus Shepard in being something that is possible to roleplay at points but ultimately not who the character of Shepard actually is canonically developed into.

Since you seem to like the word so much for some reason, a "hypocrite" is by definition one who makes a "claim or pretense of holding beliefs, feelings, standards, qualities, opinions, behaviors, virtues, motivations, or other characteristics that one does not actually hold"(<bolded part for emphasis). Thus, it is an important qualifier and cannot be dismissed unless you wish to admit that your Shepard does not actually hold  "anti AI" opinions as I and others have already elucidated, or unless you wish to change your position and argue that your Shepard is not actually a hypocrite (which would be difficult now that you have already admitted it).

Want me to leave? Find a violation in the TOS. This is a public forum, don't post in it if you just want everyone to agree with your interpretations. Otherwise I'm filing that line under QQ or "whining" as you so eloquently put. Still, at least it does prove me wrong about labeling you a troll, as if you were, you probably wouldn't be getting so mad.

 

 

You misrepresented the topic of the other thread I made. I specifically said that thread, not this thread. Please read my posts properly if you intend to reply to them. "Anti-AI" has no definition beyond "being opposed to artificial intelligence in some form or other". You've addressed maybe half of my arguments in this thread and ignored the rest. I'm not telling you how to play your game, don't tell me how I can or can't play mine. A hypocrite is also someone who "engages in the same behaviors he condemns others for" as you should be well aware, indeed making you a hypocrite for the stated reasons. Except it's more like hypocrisy squared at this point.

 

My Shepard is not somehow a hypocrite for not fitting into your narrow, rabid, imaginary definition of "anti-AI", and even if he was I wouldn't care for two different reasons. A: He's busy saving Earth, which is a lot more important than ethical or principal consistency as far as I'm concerned, and B: My Shepards aren't shallow caricatures decided for me by inconsistent dialogue trends. They're mine, and they're allowed to be hypocritical if I think it fits them.

 

Yes, I want you to leave. No, I can't force you, which is why I asked nicely. And I'm not mad, this is getting more and more satisfying the shallower your arguments become. It's not conductive or relevant to the actual thread, though.

[themikefest]

I gotta say though, I played through the third game just last week and I can barely remember what the Geth "nonsensus" mission was even about, and I have no recollection of the scene on the Normandy you're talking about.

here's the renegade response
here's the paragon response
 

How many times have you played these games?

Look on my profile

[Vazgen]

here's the renegade response
here's the paragon response

Here's the renegade response with Legion. Paragon is the same as for Geth VI, if I'm not mistaken.

[ThomasBlaine]

here's the renegade response
here's the paragon response
 

Look on my profile

 

Here's the renegade response with Legion. Paragon is the same as for Geth VI, if I'm not mistaken.

 

Coolsauce. Okay, I think the "I don't want them to die" option is noncommittal enough to make it plausible that Shepard is just mitigating between the two and trying not to antagonize the platform while it's being so very helpful. Even if it's a little inconsistent, first expressing skepticism that the Geth can be said to be alive and then referring to them as not necessarily needing to die. But that could just be Shep throwing Legion/the VI a bone after pressing a little too hard earlier in the argument.

 

I don't know about convincing Legion that the Geth need to be destroyed through the renegade options. Shepard is right enough, but him actually telling Legion this to its face and Legion accepting it reluctantly makes the whole thing weird somehow. Any thoughts?

[Vazgen]

I don't activate Legion and deal with the Geth VI. And by being Renegade with it right from the start, it makes sense for Shepard to be clear about it. With the Reapers = dead. Deal with it 

[ThomasBlaine]

I don't activate Legion and deal with the Geth VI. And by being Renegade with it right from the start, it makes sense for Shepard to be clear about it. With the Reapers = dead. Deal with it 

 

I don't know, it just feels like a weak narrative if a strange new VI shows up to help "reform" its own kind and we just freely trash talk Geth all the way through the story arc, and then it inexplicably banks their existence on us choosing not to blow its circuits out before it can turn them on the Quarians again. With Legion at least there is the vague sense of familiarity letting it think that we might give it the benefit of the doubt when the time comes, and/or enough accumulated data on Shepard's charisma and determination to unify against the reapers as to suggest that he should be able to talk the Quarians down.

[Vazgen]

I don't know, it just feels like a weak narrative if a strange new VI shows up to help "reform" its own kind and we just freely trash talk Geth all the way through the story arc, and then it inexplicably banks their existence on us not blowing its circuits out before it can turn them on the Quarians again. With Legion at least there is the vague sense of familiarity letting it think that we might give it the benefit of the doubt when the chips are down, or something.

It needs Reaper control removed. Shepard wants that too. The goals align. In the end it offers a choice and explains it "Why not? Our fleet is massive. We can assist in Shepard's fight against the Old Machines... if the creators no longer threaten us". Logical offer, it thinks. IIRC, there is no point when Shepard expresses willingness to kill the geth (the true geth, not Reaper-upgraded ones). I also don't see anti-AI Shepard ever activating Legion in ME2.

[themikefest]

Majority of the time I don't activate the geth if I don't send it to Cerberus. If I do send it to Cerberus, I have to destroy it on Chronos

[Vazgen]

I can still see that happen, simply out of curiosity. Why it talked, and why the N7 armor.

Uhm, I doubt that. Too risky IMO. And Cerberus can get that info from its databanks