Some players are rooting for the bad guy as humanity’s last hope, but why?

Warning: There some SOMA spoilers ahead.

There are lots of reasons things have gone so bad for SOMA’s underwater station, PATHOS-2, but the AI overseeing its operation, WAU, hasn’t helped. But in the eyes of some players, WAU is the one you should be rooting for. SOMA’s developers were not expecting this.

Advertisement

WAU, otherwise known as Warden Unit, works with the humans on PATHOS-2 to preserve the future of humanity. In the world of SOMA, Earth’s governments failed to prevent a comet from colliding with the planet. The plan was to destroy the comet, but failing that, PATHOS-2 was part of a contingency plan to keep some part of mankind going. When the comet scorched everything above the water, PATHOS-2’s mission became even more important; they’re all that’s left.

WAU’s main directive is to protect humans, but what happens when doing so goes wildly outside of the parameters its programmers expected?

With Earth burned to a crisp, one of the engineers on PATHOS-2 came up with another way for mankind to live on: a virtual reality simulation called the ARK. In SOMA’s world, it’s possible to copy and paste the human mind around, which is how the main character ended up in PATHOS-2 in the first place. Once everyone’s loaded into the ARK, the computer’s fitted inside a satellite that’s shot into space, left to drive through the cosmos.

Advertisement

When people started scanning their brains for the ARK, a few people viewed that as having two of themselves in the world—a contradiction. It freaked them out, so they committed suicide. WAU immediately took notice of this, and began doing whatever it could to preserve humanity. This caused a few unintended consequences. One, it started stealing brain scans from the ARK and dumping them into robots. Two, it kept other humans alive by whatever means possible, even if it meant forcing their lungs to breathe and and their hearts to beat.

WAU was partially inspired by a thought experiment popularized by Nick Bostrom, director of the Future of Humanity Institute at Oxford University, for his recent book, Superintelligence: Paths, Dangers, Strategies. (Eliezer Yudkowsky actually came up with it.) In it, Bostrom imagines an AI, developed by humans, that’s intended to create paper clips with maximum efficiency. Seems innocent, right? But where does the AI draw a line?

“How could an AI make sure that there would be as many paper clips as possible?” he told Salon in August 2014. “One thing it would do is make sure that humans didn’t switch it off, because then there would be fewer paper clips. So it might get rid of humans right away, because they could pose a threat. Also, you would want as many resources as possible, because they could be used to make paper clips. Like, for example, the atoms in human bodies.”

In WAU’s case, while its intentions were good, its execution is a bit...off.

“WAU has a purpose,” said SOMA creative director Thomas Grip. “You never know how they’re going to apply something, even with the best intentions coded. It’s very hard to see stuff that. In one situation it’s like ‘Shit, it is doing fine!’ But suddenly, if stuff happens—like the rest of the world dies—it’s directive might get a bit weird, even if it’s just following stuff. You can’t really rely on them doing choices right. It’s very, very hard to program nuances. That’s what happened here with WAU. He went ‘OK, I’m going to preserve life. What ways can I preserve life? I guess keep all these people breathing and thinking. That’s enough!’ [laughs] Oh, we need some happiness, too? Just inject the right drugs into them.”

Advertisement

Unlike SHODAN or GLaDOS, WAU is not evil. It doesn’t cackle like a supervillain; in fact, it doesn’t even talk. But while exploring the freak show that PATHOS-2 has become, it’s hard to argue WAU has set humanity on the path to prosperity.

Or has it?

The most heated discussion on SOMA’s subreddit is about this very question, in which a number of players believe the WAU is actually humanity’s best bet, and despite some things having horribly awry, it should be given more time to learn.

Lay out your argument, 1-1is0.

Why is WAU keeping everyone alive in a grotesquely mutated form?

Because WAU’s priority was saving mankind. This meant both keeping every organic human (the Primes) physically alive forever and connected to a WAU network, as well as making robots from scans. Eventually WAU would likely have been able to create actual robot bodies for the Primes themselves so long as both the original body and the robot were physically connected. Or maybe not but it is sort of a philosophical question as to how exactly the direct interface works. It may just be a form of scan too. At any rate, lots of robots like Simon was the future plan for WAU.

Why did WAU kill some of the few humans remaining on PATHOS-2?

It had no choice however, because much like WAU’s action in preventing the stupid humans from committing suicide at Theta, it had to take action to prevent the stupid humans from killing all the personalities stored within WAU and ultimately dooming mankind by killing WAU.

Why does WAU refuse to turn off the humans who become monsters?

And most troubling: WAU NEVER EVER EVER shuts down any of the running versions once they are activated no matter what unless it has no choice. No matter how demented, or insane, or crippled a running version may be, WAU considers it just as human as any prime and is loathe to kill one unless it absolutely has to. Not only does WAU consider a running version human, it considers EACH ITERATION to be a completely separate and individual human with all the benevolence and protections that this mandates. The benevolence is not absolute, but it is pretty damn benevolent, and even when it isn’t it’s not a total loss because there are plenty of backups at WAU.

1-1is0 is not alone, either.

That’s putting a lot of faith in something we can’t control!

Ooookay, good point. Maybe WAU is incapable of expressing what its ultimate intentions are to the very humans it’s trying to “protect” in the short term.

Grip didn’t expect this reaction, and didn’t write the game with this in mind.

“There are tons of people who are totally siding with WAU,” said Grip. “’The ARK is a crap project! What are you going to do out in space? That’s hellish! They should stay there and just let WAU fix everything for them. I mean, Simon turned out pretty good, he’s getting better by the day. This is gonna be fine! He’s gonna repopulate Earth! WAU is the good guy you should be rooting for!’ Personally, even if I sat and thought about it a bit, I never had that stance. [laughs] But, shit, that argument is pretty good!”

A few of SOMA’s players seem to have more faith in WAU than its creator.

As for me, screw planet Earth and screw WAU. I don’t blame WAU for the way it’s been acting, but that doesn’t mean I wanna stick around and cross my fingers. Load my consciousnesses into the computer and let’s head to the stars!

Advertisement

You can reach the author of this post at patrick.klepek@kotaku.com or on Twitter at @patrickklepek.