Advertisement

Truly intelligent enemies could change the face of gaming

Too real to kill.

Live, die, repeat -- the tagline for the 2014 science-fiction film Edge of Tomorrow -- describes its protagonist, who "respawned" every time he died in the real world. Critics noted that the conceit resembled the cyclical experience of playing a video game, in which dying resets a staged arrangement of obstacles. Often these are enemies, and the most common way they're surpassed is by the player violently dispatching them. Some games have kept this as cartoonish as Mario jumping on a Goomba's head, but others strive for vivid action and more-lifelike foes to pit the player against. But we know what enemies look like today -- how will we treat them in the games to come?

Put another way: How will violence in gaming change in the future?

The question is broad, and a little loaded. Gaming's evolution was stricken by moral panic about the effect of violent video games on kids. In those days, Mortal Kombat and Doom convinced the fearful that engaging in bloody digital combat -- as real as it looked in glorious 16-bit -- would warp players' minds. This cultural anxiety still spikes from time to time, but as those youths grew into adults no more prone to carnage than anyone else, the argument's long lost its teeth.

Freed from cultural pillories, gaming started looking inward. As the industry yearns for artistic respect, critics are asking more of violent games. In the mainstream big-studio titles, we still shoot, stab and detonate digital enemies, but some see the relationship between players and foes as ripe for exploring. One of the most popular games that ventured into this territory was Shadow of Mordor.

Released by studio Monolith Productions in September 2014, Mordor placed the player in the boots of Talion, a Ranger of Gondor who is unjustly killed. His body possessed by a wraith, Talion becomes a revenant obsessed with avenging the murder of his family, while the elven ghost inside him remembers a forgotten past. His wraithlike powers enable him to mentally dominate Orcs and set them on their former comrades. The murky moralism -- you essentially enslave Orcs but use them for "good" -- and brutal violence disturbed some reviewers, but it didn't stunt the game's success.

Monolith is gearing up to release a sequel, Shadow of War, this August. This time around, Orcs follow Talion willingly as he challenges Sauron for rule of Middle-Earth, fighting his enemies (Orcs still faithful to Mordor) and even forming a bodyguard cadre to protect their leader. Ditching mind control in favor of a strongman cult creates a more fertile arena for dynamic relationships to develop between players and their erstwhile allies and enemies, according to Michael de Plater, creative director for Shadow of War at Monolith.

"Unlike in Shadow of Mordor, where basically their whole minds got wiped by that experience, we wanted [the Orcs] to keep their personalities, and also still have some possibilities where they can still break free and even betray you or go back to Sauron. So it is more of an allegiance of a relationship rather than just straight enslaving them," de Plater told Engadget.

"There's a definite tension between those two things -- are you enslaving them, or are you recruiting them?"

"But at the same time, it is still definitely under the influence of the Ring of Power. There's a definite tension between those two things -- are you enslaving them, or are you recruiting them? And how free are they, and what sort of stories can get created with that? ... We certainly explore through the course of the story ... some of the Orcs react to, and explore, and talk about, and create different events as well," de Plater said.

Like the first game, Shadow of War will use the "Nemesis" system to generate unique Orc captains and pit them against each other in a bloody hierarchy. The system ended up creating memorable enemies that players would recall years later on Reddit forums, de Plater said: They'd kill named Orc lieutenants the system assigned a random personality and set of attributes, who would return with a grudge and sometimes kill the player, creating a brutal cycle with both parties knowing the score.

Most but not all of these named Orcs will be randomly created. Some will still be scripted by Monolith to fulfill a specific purpose in the game's story. In future games in the series -- and across the industry, in general -- de Plater hopes that procedurally-generated enemies will have such advanced AI and reaction protocols that players will develop relationships with them without realizing the NPC's lines and behaviors weren't written out beforehand.

"The ultimate goal would be that you have something that looks and feels at the level of detail of something that was entirely scripted, but it is taking place entirely procedurally," de Plater said. "I think everyone's still a fair way away from that."


The Monolith Productions team isn't alone in believing that one of gaming's frontiers lies with the unpredictability of AI-controlled enemies and allies. Mitu Khandaker teaches on the topic as assistant arts professor at NYU Game Center -- but as chief creative officer at artificial-intelligence company Spirit AI, she's also working with a team to develop technology for companies to use in their own games.

"What we do is build tools to help developers creatively author story scenarios and author personalities for characters and the kinds of things that characters might say, but then those characters might improvise based on the space that you've authored for them," Khandaker told Engadget. "There's a lot of potential there for players to really have deeper, more meaningful conversations with characters."

"There's a lot of potential there for players to really have deeper, more meaningful conversations with characters."

Spirit AI's efforts could be summarized as "building technology which will let us make the walking simulator a conversation," according to Khandaker. Think of the squad's idle chatter in Mass Effect, or the casual smalltalk during long car rides in Final Fantasy XV: Pre-written, nonessential dialogue tumbling out of an algorithmic generator that organically delivers exposition and character detail. But what if those AI characters talking to the player and making up responses on the fly — even if they're enemy grunts with their guns drawn?

Khandaker can imagine creating games where the enemies aren't just tokens or pawns but more fully formed virtual characters. "Instead of just committing violence upon some kind of enemy, maybe [players will be] trying to understand their motives, she said. "Now, in this cultural context, more than ever, a human understanding of the reasons why people make decisions they do is super-important. Even if, on some level, we think decisions people make might be evil, we still need to have the level of understanding because that's how we learn and grow and how we combat evil."

What Shadow of War won't have are human enemies that players can mind control or kill in gruesome ways: Your foes will be Mordor-born Orcs who span the gray-brown gamut and exhibit the violent, traitorous ways of their race. This is intentional.

"One of the challenging things is striking the balance of having a game that's fundamentally pretty gritty and violent, but also making sure that we have this humor in there and this levity to it," de Plater said. "Ultimately, even though it is dealing with some dark themes, there is a cartoony level of violence as well. Orcs represent these caricatures. Everything's turned up to 11 in terms of their personality and their characters and their faults, and the violence of their society and how power-crazed they all are; how backstabbing and cutthroat they are against anyone."

In short, you'll be dispatching and commanding a class of enemy designed to be dynamically interesting yet disposable in a way that shouldn't trigger a player's ethical qualms. Game critic Austin Walker believed that the first game, Shadow of Mordor, failed to justify Talion's anti-Orc kill-and-enslave crusade: "But we're told again and again that these Orcs want to destroy beautiful things. It just doesn't hold up, and this tension extends to every element of their narrative and systemic characterizations. These Orcs have fears, interests, values, rivalry and friendships. Some Orcs are lovingly protective of their bosses or underlings. But they are 'savage creatures' that 'hate beauty,' so go ahead and enslave them," Walker wrote.

At least Shadow of War will strive to explore new and uncomfortable relationships between player and enemy. Even if it never lets players forget Orcs are villains at their core, some will attempt to liberate themselves from any overlord, dark or bright, de Plater said. He didn't specify whether these autonomy-seeking enemies will be a scripted faction in the game. But imagine wandering down the sludgy Mordor foothills only to find a procedurally-generated band of Orcs that avoid conflict and try to run away from you, the bogeyman who's murdered (or recruited) all their friends, as they search for a better life.

Imbuing enemies with relatable traits -- human traits -- is as fascinating as it is discomforting. Since their inception, single-player games have driven a hard wedge between players and enemies by making the latter alien and threatening. Space Invaders and Galaga literally used aliens, while Missile Defense tossed unthinking explosives at the vulnerable people populating the player's cities. The dawn of the first-person-shooter genre featured demonic monsters in Doom and Nazis in Wolfenstein 3D, enemies so unrelatable that players don't think when gunning them down.

Spirit AI's clients are using its AI-conversation tech to augment NPC allies, though Khandaker's team is starting to graft it onto enemies. But it's really up to whoever uses Spirit's tools, and whichever studio decides to challenge players with ordinary foes that do more than shoot in their direction.

"I would love to see that as a moral choice that you make. It should be sometimes deeply troubling, depending on your particular game, that somebody is so human and so full of their own motive, doing the things that they're doing, that it's not so easy to dehumanize them," Khandaker said.

"I think that through good, well-considered design, we'll get to a point where actually these interactions with characters help us to better understand the motivations that real people have."

"This is why I think it comes down to designing photo-realistic, naturalistic AI really well. If [designers] let you push them around, you're going to maybe transfer that to real people. If, however, they don't — if they push back and they try and do the emotional labor of helping you to understand what it is to interact with someone in a nice, well-considered way — then you can maybe transfer that to your interactions with people," Khandaker said. "I think that through good, well-considered design, we'll get to a point where actually these interactions with characters help us to better understand the motivations that real people have."

Whether AI tech will develop substantially in the next few years and, ultimately, whether improving enemy and ally AI will positively affect the player's experience, is another question. As Compulsion Games' Creative Director Guillaume Provost points out, making smarter enemies doesn't matter much if the player doesn't know what's going on.

"Making AIs that are believable often involve stuff that's not that technical and has a lot more to do with the acting parts that are involved in the AI," Provost said. "So it's not so much the sophistication of the technology behind it as it is the sophistication of expressing what's going on in their heads to the player."

"It's not so much the sophistication of the technology behind it as it is the sophistication of expressing what's going on in their heads to the player."

For Provost, that meant tweaking some gameplay in Compulsion Games' latest title, We Happy Few, which was released in Early Access last year. In it, players try to escape an English city whose denizens imbibe drugs en masse to forget their communal crimes -- and punish those who won't do the same. In playtesting, this meant making the hostile NPCs warn the player several times before violently reacting. They couldn't assume players would pick up on cues because in gaming, players' attention is focused on what they're interacting with at the time.

"The truth is, it's not a movie where you sit down and watch people the whole time. You're actively doing stuff. You're running around, you're stealing stuff. The player has a smaller portion of their brain left to understand what the people around them are doing," Provost said.

Which is why developers have to treat player attention as a resource and be smart about what they make intelligent. Provost recalled a story about the grunts in the first Halo who were programmed to yell out "I surrender" and wave their arms around -- but players would gun them down before the little enemies could bark out their lines. Similarly, Provost doesn't see nearly as much use for plugging more AI into enemies to make them smarter in future games.

"The biggest advances that I found in my actual experience don't have that much to do with technology. They have to do with the sophistication that we can actually window-dress AIs as human characters. And there's nowhere we've moved for faster in that area, I think, than in companion AIs that are going to accompany you, and for the simple reason that you have much more opportunities to actually build an emotional response to those people," Provost said. "If you walk down the street and shoot a guy, you've literally spent 10 seconds with him."

Which explains the success of companions like Elizabeth in BioShock Infinite or Ellie in The Last of Us. Specifically, their AI was deliberately programmed to recognize downtime and start chit-chatting -- the smalltalk between action set pieces that draws the player closer to companions. But players don't spend nearly as much time around enemies, so there's much less opportunity for their programming to shine. And if you plug AI into their performance, well, it's really easy to make an AI that never misses -- that's why cheaters in shooter games use aimbots.

"The biggest fun there is in AI is being able to predict successfully what they're doing, which is completely counterintuitive to the idea of having this really deep AI. If you're playing a game and the AI's always completely unpredictable, it just turns into a frustrating experience for the player because they can't learn a good strategy to actually succeed at the game," Provost said. "[The key is] having a good balance where, over time, the player gets to master a game by understanding, whether at a conscious or subconscious level, what is likely going to result from their action and being able to strategize which actions to take. It's a cornerstone of making AI interesting when you're playing them as foes."

"If you're playing a game and the AI's always completely unpredictable, it just turns into a frustrating experience for the player because they can't learn a good strategy to actually succeed at the game."

There's an obvious question here: Why make enemies more complex if you're just going to shoot them? Players have been dispatching foes since video games moved past Pong 45 years ago. They are obstacles. But big-budget studios are spending lots of money making them look like really pretty obstacles to shoot at. A time might come where the disparity between human-looking-but-robotic-acting enemies becomes too jarring.

"It's a problem when you're making games increasingly photo-realistic and though the enemies are representative of real people, they don't really act like real people. If enemies are just cartoonish representations, then perhaps it's not a problem if they're not fully-formed individuals," Khandaker said. When enemies are photo-realistic and representative of gender or race or a demographic, it's more problematic to blur identities. "You see this a lot with games representing terrorists. It's often this idea of 'generic brown person.' I think that is a problem. We need to start understanding that people who look like humans should also behave like humans."

In 2017, imposing greater sensitivity in games to stay away from stereotypes seems like a no-brainer. Not that games aren't still committing the sins of racial omission, but the cure is obvious: diversity and research. Striving toward accurate representation.

Whether to make enemies smarter is another question. Especially when big-budget games are crafted to be enjoyable experiences that don't tax your brain. They spur a pleasurable feedback loop, said Miguel Sicart, associate professor at the Center for Computer Games Research at the IT University of Copenhagen.

"By the end of the day, what players want is to have enemies that have patterns that are ultimately recognizable so we can beat them. That's the goal of every player: to get so good at something that they can beat the enemy," Sicart told Engadget. At their core, a lot of big-budget games aren't really designed to think deeper than providing satisfying player-vs.-computer combat. "We are never going to see Call of Duty be morally nuanced, the same as we are never going to see The Fast And Furious be morally nuanced; but they also don't need to be. They're just empty popcorn that feels really good."

The violence in many games -- especially ones putting the character in military boots gunning for enemies of freedom -- doesn't bother Sicart. It's the lack of context that worries him. Why are players where they are, mowing down digital interchangeable mooks? Are there consequences to going to war?

"There's currently a rhetoric [in gaming] about war that's not glorifying it, but not necessarily not glorifying it," Sicart said. "You can see this in the US media when Trump ordered strikes on Syria and the media went crazy about the beauty of these missiles. We've lost a little bit in the media landscape -- this capacity of saying, 'Holy shit, we are throwing missiles at people.'"

That disconnect troubles Sicart. People uploading videos to YouTube of gruesome kills and testicle-shattering sniper montages is a glorification of sorts, which is "one step toward this trivialization of war and conflict and death and violence," he said. We've always had violence, stretching back to the Iliad and Odyssey graphically describing dismemberment-by-cyclops, Sicart points out, and violence is a part of how we express ourselves as humans. But a lot of war games make violence the unique selling point: They are about the carnage. They are about the death.

"... Some of these games are propaganda. To me, they are the worst type of propaganda because it's propaganda that you not only consume by reading and listening, but you have to participate."

"We have to be careful with how we sell what, by the end of the day, is just coppers and robbers on a computer. Careful how we sell these things and careful with how we try to tie it to real military discourse. Careful how we try to tie it to conflicts in the real world. We have to be careful with these things, because we are feeding a general media discourse that maybe we don't want to feed," Sicart said. "I mean, to be perfectly honest, some of these games are propaganda. To me, they are the worst type of propaganda because it's propaganda that you not only consume by reading and listening, but you have to participate."

The point isn't just to be careful about what studios are making for their audiences, but how aware they are of their medium's failings. The 2012 cult classic Spec Ops: The Line has been duly lauded for its self-awareness amid a sea of identical military shooters (what Sicart calls "these brown-looking video games where you are a disembodied gun shooting other people") for pushing players to seriously evaluate why they thoughtlessly careen through other titles gunning down whomever is tossed in their way. But that was five years ago. Since then, which games have come along to seriously challenge the player's relationship with violence?

"It may take five years, or maybe 10, but I think we are slowly getting to the position when video games are wanting to actively participate in the cultural debate, and therefore we will have a much more nuanced take on violence," Sicart said. "We will be able to call bullshit on violence for violence's sake. We will be able to call bullshit on video games that just want to glorify discourses that we don't want in culture."


That's the same argument we've been hearing for years. But video games are slowly opening up to the increasingly widespread format of virtual reality. With top-line headsets getting bundles and smartphone-powering headsets like Google Daydream and Samsung VR getting more support, the medium is growing day by day. Its tech immerses players deeper into the worlds and experiences developers create for them.

Khandaker's doctoral research involved putting test subjects through the paces of a game she made. Players went on a climb with a friend, simulating the tactile hand-over-hand action of ascending a rock face grip by grip. You reach the top ahead of your partner, when she suddenly drops, dangling far below on a rope rapidly fraying under the weight of both of you.

"This NPC starts screaming at you to just cut the rope and just let her die, and all of these things. As the player, you have to make the decision about whether you do that, and it's very time-sensitive, so you have to make the decision about whether you're going to cut this friend of yours loose and ultimately murder her, or are you going to not make a decision and let the rope break and kill the both of you," Khandaker said. "It was really fascinating because the decisions that people made were very different in VR versus as a classic interface. People took longer to make the decision about what to do, people felt worse about it, in VR. People felt closer to the character in VR."

"It was really fascinating because the decisions that people made were very different in VR versus as a classic interface. People took longer to make the decision about what to do, people felt worse about it, in VR. People felt closer to the character in VR."

But after a follow-up questionnaire, Khandaker sat down to interview every player, and most of them -- even the ones who were visibly upset -- settled down and told her it was fun, that it was just a game. Even in the immersion of VR, players retain the double consciousness of emotional involvement in the experience with awareness that it's still a simulation.

"That's always been there, even if you are doing the worst kind of dehumanizing-- sort of mowing through enemies in a photo-realistic game. You can slightly feel bad about it, but you also know it's a game," Khandaker said. "In VR, that effect is even more pronounced. We feel worse about the things we do but we also know what we're doing is just a game. There's a lot of potential to play with those feelings. Especially ... if characters seem even more like real characters, if they're able to respond to you."

In other words, VR immersion augments gaming, but it's the connections that dramatically affect the player experience. Part of that means simulating human conventions -- move this way, react that way, emote like we would -- softening their perfect reflexes and senses to let players compete on more even ground.

"Making an AI really good or really skilled, that's a problem we solved day one. We spend an inordinate amount of time making them suck. Making an AI look really smart, because they're either flanking you or they're appearing to work as a group together, that is immersion," Provost said. "It's not intelligence, and the immersion is done by us collectively understanding what are the best ways to communicate to the player, 'Hey, I'm doing something intelligent,' whether they're really doing something intelligent or not."

"Making an AI really good or really skilled, that's a problem we solved day one. We spend an inordinate amount of time making them suck."

For Eric Zimmerman of the NYU Game Center, however, humans are still the apex of depth and complexity. We'll probably get complex character moments from AI partners in the future, but the best examples are in human-to-human interaction, Zimmerman maintained. Skirmishes, deceptions, alliances built up over time only to get dashed by a mole intent on starting a war -- you can find that, and more, in EVE Online thanks to its dedicated player community.

"The stories coming out of EVE Online are the stories people want to come out of other games," Zimmerman said. Most recently, the space-age MMO saw the leader of one of its most elite pirate gangs get conned out of a one-of-a-kind ship worth 300 billion in-game Isk (or about $2,600 in real currency, according to this exchange). But the game is infamous for its betrayals, faction infiltration, frontier piracy and colossal space battles, all happening at a scale that would seem impossible for a single game filled with AI NPCs to mimic.

Human interaction or intention in design will bring depth, Zimmerman concludes -- not purely new technology. He recalls a moment during a presentation at a game developers conference in the early 2000s when a Sony executive was crowing about the latest cutting-edge development in the then-latest PlayStation console. "'Finally, we'll have high-resolution tears running down high-resolution cheeks as they're crying, and we'll finally have deep emotions in characters,'" Zimmerman recalls him saying.

"The idea that technology guarantees a meaningful experience is false. All of this tech is still just an expressive tool for people to use to express themselves, to express ideas, to make statements about the world," Zimmerman said. "The human decisions that go into a game are far, far more important than the technology that's driving the experience."

But it's the awkward attempts, the false starts and grand aspirations dashed by bizarre results that mark the progress of video games. Ambitious games that failed to live up to their hyped AI complexity, like Peter Molyneux's Black and White or Stephen Spielberg and EA's scrapped LMNO project, nonetheless leave conceptual scraps for later titles to pick up.

Among the disappointments are successes like Left For Dead, which debuted in 2008 with its "AI Director" that shuffled enemies and items around for dynamic and dramatically different playthroughs. Half-Life 2 debuted in 2004 with its AI companion Alex, and almost a decade later in 2013, gaming got her successors Elizabeth in Bioshock Infinite and Ellie in The Last Of Us. Imagine what kind of AI-filled worlds we'll get to play in a decade from this August's Shadow of War.

Image credits: Matthew Lyons (lead illustration); Warner Bros Interactive Entertainment ('Shadow of War'); 2K Games ('Bioshock'); 2K Games/NeoGAF ('Spec-Ops: The Line' screenshot).


Welcome to Tomorrow, Engadget's new home for stuff that hasn't happened yet. You can read more about the future of, well, everything, at Tomorrow's permanent home and check out all of our launch week stories here.