Although you might think of things like self-driving cars or chatbots, AI in gaming has quietly become an integral part of the industry. From improving graphics to creating lifelike characters, AI in gaming has significantly changed how we make and experience video games. As technology develops, AI in gaming promises even more exciting advancements in how games look, sound, and play.
Some predict AI may even drive the gaming industry to reach its true potential. To understand just how important AI is and will continue to be to the industry, we must first look back at how far things have come.
Table Of Contents:
- AI in Gaming Impact on How We Play Games Today
- 8 Ways That AI Could Reshape How We Play Games in the Future
- 1. Natural-Sounding Conversations
- 2. Infinite Possibilities for Open World Gaming
- 3. Generating Dialogue Based on Player Actions
- 4. Improving Cloud-Based Gaming Integration
- 5. Enhancing Full-Body Motion Tracking for VR
- 6. Using Procedural Level Design
- 7. Predicting Player Actions and Designing Games Based on Them
- 8. Dynamically Changing Game Difficulty
- The Challenges of Developing Responsible AI Technology
- FAQs About AI in Gaming
- Is AI Good or Bad for Gaming?
- Conclusion
AI in Gaming Impact on How We Play Games Today
When playing visually complex games such as Minecraft or God of War, with their intricate worlds, graphics, and vast landscapes, you might think sophisticated algorithms drive the game’s actions.
Game AI today works by taking images frame by frame and modifying them inside the game. For example, generative AI in gaming takes real-life images of cityscapes to help create photorealistic landscapes. These landscapes look just like the stunning views you see while racing around California in Grand Theft Auto 5.
Advanced AI in Gaming Concerns
Many video game developers are holding back from using advanced AI in gaming, not only because of the cost. Developers worry advanced AI in gaming may lead to a loss of control over the game-playing experience. While games can appear very complex and involve several nuanced layers, their outcomes must be somewhat predictable by nature.
However, as Greg Brockman, head of research at OpenAI, explained in an interview with the New York Times, AI can do much more. If AI can be taught to beat games, it can problem-solve for anything, such as improving AI applications related to manufacturing, finance, or healthcare.
Although it seems counterintuitive that teaching an AI to win video games could solve important real-life problems, many research labs are experimenting. They’re using games to advance AI technologies that could ultimately improve lives. A good example of this is DeepMind.
Scientists at DeepMind use a trial-and-error process called reinforcement learning to train AI in gaming to problem-solve. They’re teaching it how to play simple games like Pong. Eventually, those AI will graduate to complex games like Starcraft 2 and Dota 2, hopefully enabling AI in gaming researchers to transfer those skills toward developing technologies to solve bigger challenges across industries.
Smart Opponents and Better Games
So, is this to say developers have given up entirely on improving our gaming experience? Definitely not. AI in gaming today is primarily used to create machine learning systems. These systems enhance computer graphics and develop ‘smart’ opponents to make games more fun. The ability to match gamers from around the globe using broadband and 5G ensures there will always be challenging opponents for anyone who wants to take on other humans. But many expect this will soon oscillate back to developing smarter AI-generated opponents, at least eventually.
But there is another factor driving AI in the gaming world: the massive increase in simultaneous players and an increasingly connected world. Gaming isn’t limited to consoles or PC games anymore; with cloud-based gaming, players expect immersive and fast-paced games across many platforms, even mobile. The focus now is not just to provide enough content to play. It’s to create personalized and engaging games across devices, with AI in gaming technology driving the backend and development, helping make it happen faster and at scale.
This means that developers are working on training AI in gaming systems. These systems focus on analyzing player behaviors to develop even better user experiences that tailor gameplay and the game’s speed to our personal skill levels. In many games today, AI can even help predict how likely a player is to stop playing or what makes gamers quit certain games altogether.
8 Ways That AI Could Reshape How We Play Games in the Future
As AI technology advances even further, developers are excited about the numerous innovative ways that it will revolutionize gameplay. The potential to change how games are made and enjoyed, providing more immersive, personalized, and challenging game play, is enormous. But to fully understand why, it’s important to understand how the future of gaming could be different from what we experience now.
1. Natural-Sounding Conversations
Today, the stilted and repetitive speech and gestures of most non-playable characters or NPCs often detract from a truly immersive experience. We’ve come to expect NPCs to react robotically when you interact with them. Until recently it hasn’t been possible to create NPCs who feel or behave like human players.
Generative AI tools, such as La Forge’s Ghostwriter, can make dialogue unique and give NPCs much greater variability in what they say to a player. AI may finally be the key to making those conversations more convincing because it helps to develop a vast catalog of responses and unique speech patterns to respond to any number of actions or triggers.
The video game company Ninja Theory uses generative AI to generate vocal performances for NPCs using the Altered AI voice library.
2. Infinite Possibilities for Open World Gaming
Until recently, most game developers relied on pre-scripted content, often choosing from hundreds or thousands of dialogue options to add interest to NPC conversations. As anyone who’s enjoyed playing games like Skyrim can attest, although seemingly endless, even the most robust games with a virtually limitless playing field still have limitations.
However, as Julian Togelius, an AI researcher at NYU, pointed out, games like No Man’s Sky allow players to enjoy what is truly an open-world experience without having to replay the same storylines. Imagine 18 quintillion different variations of a planet to explore. Using procedurally generated content, developers can achieve this “virtually endless game,” as Togelius describes it. That means customized landscapes, plant life, animal life, and even the composition of the atmosphere.
By taking some pressure off of writers and designers to think up every detail of the environment, procedurally generated content allows AI to take over. Developers can spend their time improving other aspects of gameplay. This also gives players truly unique experiences. Think of the enormous possibilities.
3. Generating Dialogue Based on Player Actions
“Detroit: Become Human,” allows players to make choices dramatically affecting gameplay and story arcs. Each playthrough can result in completely different interactions and even different storylines. Although “Detroit: Become Human” was designed to include countless different branching pathways using traditional writing techniques, the ability to achieve such a feat using current AI capabilities opens up exciting possibilities to other studios.
Imagine an adventure game in which your character encounters completely new scenarios based on how they react to situations in-game. While it sounds improbable that AI will someday be capable of writing in the same way as an expert storyteller, using natural language processing could eliminate the need to plan every possible outcome.
With the potential for advanced large language models, like those currently being tested and developed by Microsoft and Google, such a feat is almost inevitable. It may even enable developers to build gaming scenarios in real time. That kind of creativity and fast-paced development is exciting for developers. Microsoft seems to agree with this because it recently announced in November that it’s working with Inworld AI to Develop AI tools that will produce more engaging and natural-sounding narrative and dialogue tools.
4. Improving Cloud-Based Gaming Integration
In many ways, the future of gaming depends heavily on making cloud-based platforms as immersive and functional as a traditional console. As anyone who has used cloud-based gaming systems knows, lag and buffering issues can cause big problems that ruin gaming enjoyment. Fortunately, with machine-learning algorithms, this may soon be a thing of the past.
As broadband and 5G technologies advance and AI algorithms continue to improve at analyzing our gaming data and behaviors, game developers will eventually be able to optimize gaming servers using AI tools to tailor those systems to each gamer’s platform.
Also, developers could soon have the tools available to develop far more complex games with no loss of game performance for even mobile or tablet devices.
5. Enhancing Full-Body Motion Tracking for VR
AI-assisted full-body motion tracking will take VR game experiences to another level entirely. This technology uses video game development algorithms to track players’ physical movements and seamlessly translate them to gameplay in virtual reality.
We saw how exciting full-body tracking might become with Xbox’s now-retired Kinect. Though the technology was amazing at the time, Kinect’s dependence on pre-set instructions limited its full potential. As processing speeds and AI technologies advance, there’s no reason why advanced full-body motion-tracking games may soon be enjoyed across all VR platforms. This could give players the ultimate opportunity to engage in game-playing scenarios that truly feel real.
6. Using Procedural Level Design
Gone are the days of manually creating each level. This frees developers up to design new gaming scenarios while providing far more intricate landscapes that can change whenever someone replays them. Togelius explained it well in his interview when describing the use of procedurally generated content. He called “No Man’s Sky” a ‘virtually endless game’ where each planet features a combination of customized geology, flora, and fauna.
| AI Innovation | Description |
|---|---|
| Procedural Level Design | Uses AI algorithms to generate and adjust game levels based on player behaviors and inputs, including customizing landscapes, plant and animal life and atmospheric conditions, creating an almost endless and wholly unique experience. |
| Natural-Sounding Dialogue | Provides natural-sounding speech patterns, conversations, and tone for non-player characters (NPCs). Removes pre-scripted responses and dialogue to allow for realistic and diverse character interaction, making conversations flow as if players are communicating with another human. |
| Full-Body Tracking for VR | Improves VR capabilities to include AI-assisted full-body tracking, removing the limitations of previous systems like Xbox Kinect to allow seamless body motion recognition, potentially creating games that could realistically emulate real-world activities, leading to much more realistic game immersion. |
| Adaptive Difficulty Adjustment | AI systems continually assess and adapt to player behaviors, skills, and experience. Modifies gameplay to match skill levels. Adjusts difficulty levels, creates challenges that meet players at their individual level of competence, ensuring players have challenging gameplay without frustration, and may help increase player engagement. |
Another intriguing use for procedurally generated levels is that it can potentially help to enhance video game guide writer tasks. Many players who find procedurally generated levels particularly challenging depend on video game writing jobs created by others to get advice on beating games.
7. Predicting Player Actions and Designing Games Based on Them
Machine learning models have gotten very good at predicting what makes humans tick. But how would that play out in game design? The beauty of AI systems that are programmed to collect massive amounts of gamer data is that they don’t need our input to form valuable insights.
While developers face some challenges in achieving ethical and responsible AI integration while simultaneously offering new products to satisfy consumer demands for unique experiences, there are already games that depend heavily on analyzing gameplay data.
What they learn from observing and assessing player behaviors is being used to build exciting new products and levels, helping developers fine-tune their offerings, helping games like the recent “Justice” mobile release from NetEase, which uses ChatGPT to generate fresh NPC dialogue. At the same time, Replica Studios introduced their innovative “AI-powered Smart NPCs.”
8. Dynamically Changing Game Difficulty
Imagine a game where AI can learn and predict how good you are or even if you might stop playing altogether. Although it seems a bit creepy that algorithms will soon have access to that kind of data, we may expect that our generative AI assistants can provide much better-tailored gaming options, free from frustration. No longer will games need to offer standard pre-set difficulty levels because adaptive AI tools will ensure a much more adaptive experience.
This could improve player engagement and create better gameplay experiences, and AI tools may soon give developers a huge competitive edge in a market increasingly dependent on personalized services.
The Challenges of Developing Responsible AI Technology
While integrating AI into gaming opens up exciting new possibilities for development, designers, developers, and ethicists face several key ethical challenges as the technology advances, specifically on the impact AI tools may have on player behavior.
AI and Electricity Use
The energy consumption necessary to sustain such massive processing powers required for even relatively basic AI has prompted some concern. Game designer Mike Cook questions the real need for integrating AI just for its sake, reminding us that although technology can make tasks easier for human developers, it comes at a real-world environmental cost. In an interview for the BBC, Cook shared his concerns when he mentioned the electricity needs for AI technology.
Although cloud gaming helps reduce some of the environmental costs of building powerful graphics cards for personal PCs or gaming consoles, Cook makes a compelling point. He suggests it’s an issue to continue thinking about.
Ethical Considerations and Challenges
There’s no denying the potential for bias in any system designed to collect, process, and analyze massive amounts of player data. Researchers and scientists now face the question of how to protect user data while simultaneously offering personalized content. This is especially true in open-world platforms or multiplayer games where hundreds of millions of users will potentially contribute toward generating a constant and neverending stream of information.
Mitigating potential risks to player privacy is of chief concern as new algorithms are trained to offer an increasingly targeted and personally curated user experience. The recent law passed in the EU attempting to reassure both government agencies and regular citizens about concerns about AI’s future uses is a good example of how these conversations are becoming important to the public as technology advances.
It’s not just regulators getting involved, though. Researchers and scientists are also thinking critically about these same problems, as evidenced in this study by Mike Orkin on AI ethics. He studied gameplay and interactions between players to learn how those could translate to creating more robust non-player characters.
Impact of AI on Gaming Jobs
Is AI a threat to employment in the video game industry? There is concern that although AI technology can certainly help speed up game development time and even free up employees to be more creative, using AI tools comes at the expense of some of those same jobs, like game artists who are hired to produce unique character models and NPCs, voice actors, or quality assurance testers who may be made redundant as new AI platforms and bots, as proposed in this paper on machine-learning game development from modl.ai, begin to outperform their human counterparts.
But will studios completely replace real people with AI-driven alternatives? That’s doubtful, particularly in roles like video game writing jobs that require complex narrative building, understanding human interactions, and subtle writing nuances that simply cannot be fully reproduced by an AI.
Jade Raymond, head developer at Ubisoft, the studio behind “Assassins Creed,” acknowledged this potential for AI’s role in developing blockbuster games, noting how helpful the technology can be when trying to create a budget-friendly and rapid game design.
Her studio already understands the tech’s capabilities because Ubisoft developed generative AI software called Ghostwriter. Ghostwriter is being used to build NPC responses and actions they’ve named ‘barks.’
Develop and use Generative AI
This trend of developers choosing to develop and use their generative AI is also evident at studios like the indie firm Hidden Door. Hidden Door is busy building a closed alpha version of their new experimental game designed to actively generate fresh storylines, quests, and characters using player input to push forward narratives, which certainly proves how influential the technology will become.
Developers aren’t just making their versions of ChatGPT either. Companies such as Electronic Arts rely on their AI tool SEED to make powerful NPCs who can learn by imitating the very best players.
And as an added cost-saving measure, using text-to-speech generators like the AI library Replica Studio has designed that allows game makers to generate NPC speech using an AI voice actor is very likely going to influence the direction game developers go in the future.
Copyright Issues
As anyone in the gaming industry knows, securing intellectual property rights can be an enormous hurdle to overcome when bringing new games to market. But when AI-generated products begin hitting the market, a new question emerges. Who owns that AI?
Although still largely unproven legally, and already the subject of ongoing lawsuits that haven’t been completely resolved, how much a company can protect the use of any specific AI from copyright issues will certainly continue to be debated. That developers and larger gaming studios alike are enthusiastic about AI’s ability to speed up design and produce personalized game-playing opportunities has led many to automate content generation, particularly if those generative AI tools allow companies to avoid paying designers and writers.
Although several of these cases remain unresolved, using existing IP without a clear legal framework, especially given concerns over how large amounts of player data might be misused or shared, could set the stage for potentially big legal and ethical challenges for any gaming studio thinking of adopting generative AI tools in development.
FAQs About AI in Gaming
How Is AI Used In Gaming?
AI in gaming can take many forms. You might see it in action with “smart” non-player characters, procedural generation of levels, or AI and machine learning algorithms used to predict game design trends or in machine learning systems to personalize user experiences or increase engagement.
Gaming companies like EA and Ubisoft have developed their in-house generative AI that relies on those kinds of gaming industry standards for creating games. Game studios use this data to learn from players and produce better products. And since these new products often give players unique gaming scenarios, it may make beating games much more difficult. Some people find AI-driven games much harder to win.
Which AI Tool Is Used In Gaming?
Many different kinds of artificial intelligence technologies h dramatically improve both how games are made and enjoyed. You might see this reflected in the improved visual quality of modern games like “Grand Theft Auto 5” and its amazing photorealistic landscapes. Or you may notice the shift from old-school, static, NPCs with simple dialog to new types of characters in games like “Detroit: Become Human.” In these games, dialogue trees, facial expressions, and movements are generated by AI and based on how we as players engage with them.
AI technologies will also soon help enhance the machine learning development and enjoyment of VR and cloud-based gaming systems. This will help them work faster, lag less, and function even better with our platforms, whether using a gaming console, personal PC, or even a mobile device.
To achieve that though, requires AI developers and game studios to rely on powerful generative AI systems that have access to large databases that store player actions and choices. As you can see, the sheer scope of these kinds of operations necessitates developers to come up with innovative strategies that use many different kinds of AI tools. It’s the combined impact of all of these AI tools, each focused on a different area of development, that make generative AI so important to game development.
How Could AI Disrupt Video Games?
AI is perfectly suited to make video games even more entertaining than ever by using our player data to fine-tune games to make them so exciting we won’t ever want to stop playing. The trend of offering more in-game microtransactions and season passes or using loot boxes is something most developers have been increasingly enthusiastic about, as any modern gamer can tell you. AI may eventually make such sales pitches irresistible.
Although no evidence has suggested that video games cause people to make addictive choices or overspend to beat them, the industry has been steadily shifting away from simple, straightforward gaming models. Instead, it is opting to give users seemingly neverending new goals, items, or quests to accomplish. Many, including Julian Togelius, question just how far the line of giving consumers more tailored experiences could go.
Will AI Make Game Development Faster?
Absolutely. AI has helped make video games more appealing for both developers and players. To put the scope of the effort in context, some of the bigger video game companies employ over a thousand artists to draw backgrounds, model 3D images for games, and design virtual spaces. Because game worlds continue to become more elaborate, hiring enough people to make content can take months or years to produce.
But for video games to stay fresh and engaging, studios have to figure out a way to offer those kinds of complicated worlds in a faster and less labor-intensive manner. Studios like Keywords Studios think AI might be just the solution to their needs.
At a recent conference, they explained how Project AVA used tools from four hundred different artificial intelligence startups to design a civilization-building game. Although not meant for public release, the project demonstrated what developers can achieve using these tools to build new products at scale.
Although it sounds impressive to say developers may someday be capable of writing five hundred thousand lines of unique dialogue, as Rockstar reportedly required to release “Red Dead Redemption 2,” according to Vulture, that doesn’t guarantee success. The key moving forward may involve blending human creativity with AI-powered capabilities rather than depending on those tools to make money quickly.
Is AI Good or Bad for Gaming?
As with any rapidly emerging technology, we just don’t know yet what the lasting impact might be. But this we can be certain about: The trend toward procedurally generated storylines and virtual environments in gaming powered by generative AI tools, combined with personalized experiences designed to meet the desires of 2.5 billion game players from around the globe suggests AI in gaming will likely become as ubiquitous as electricity is to running your console.
Will AI be able to perfectly reproduce complex narratives with realistic characters or worlds that perfectly imitate our reality? Maybe, but until then, there is no substitute for good storytelling.
We don’t need an AI system to tell us that there’s no way of knowing for sure. Only time will tell, because AI in gaming could very well usher in a bold new future where those limitations fade into pixels on a screen.
Conclusion
AI in gaming has evolved considerably, revolutionizing game development and user experience. From smarter NPCs to dynamic worlds and optimized gameplay, AI is enhancing realism and engagement. While ethical challenges regarding copyright and sustainability remain, AI continues to reshape gaming, offering limitless potential for more immersive and personalized gaming experiences in the future.
Subscribe to my LEAN 360 newsletter to learn more about startup insights.
