AI + games: is there a future?

The gaming industry has long been at the forefront of technology: graphics, physics, network code, animations, all of which were born in games and often later transferred to other industries. Now, artificial intelligence is bursting into this ecosystem, and developers, producers, and gamers are faced with the sacramental question: will AI change games as a product and a business, or will it become just another fad?

The Data Science UA‘s answer is yes, AI has a future in games. But it’s not “AI will replace game development”, but rather “AI will become a tool that accelerates, expands, and sometimes deepens” game development. It all depends on how people know how to use it or want to use it.

Below are the arguments in favor, real-life cases, and important caveats.

Where AI is already delivering value: real-world applications

– Rapid content generation (art, assets, levels)

According to a Unity report, approximately 62% of studios already use AI tools during the development phase: for prototypes, concept art, object generation, and game worlds.

In other words, AI does not necessarily do the entire project – but it speeds up routine tasks, freeing up the team’s time for gameplay, design, and creative solutions.

Tencent’s new tool, VISVISE, shows that AI can reduce days or even months of work on art and 3D models to minutes.

This approach is especially useful for indie studios and small teams, who previously had to choose between a long cycle or limited quality.

– Improved game design, dynamic worlds, and adaptive content

Today’s AI is not just an image generator, but a “brain” capable of responding to player actions, adjusting the plot, and generating dialogues, quests, and NPCs.

For example, PANGeA research demonstrates how LLM models can automatically generate plots, environments, NPCs, and scenarios for turn-based RPGs, while taking into account NPC personalities and world logic.

It makes it possible to create games where each playthrough is unique, where the world “lives” and reacts, rather than simply reproducing static templates.

AI can also bring NPCs to life, not with dry, pre-set dialogue, but with lively, adaptive interaction that dramatically increases immersion.

As a result, AI helps make games more flexible, lively, and personalized, not just beautiful.

– Automation and optimization of teamwork

Many routine tasks, creating asset variants, prototypes, minor edits, bug fixes, and build tests, can be delegated to AI. It reduces the workload on specialists and speeds up the development cycle.

A 2025 study shows that nearly 90% of developers now use AI agents to automate tasks. AI acts as a “boost” for the team: less routine work means more creativity.

– Large-scale live content and updates

For gaming projects with frequent updates, live services, and microtransactions, AI provides a competitive advantage. Rapid generation of assets, texts, quests, and adaptive balance, all of this becomes realistic and economically viable even for medium and small companies.

Companies that previously could not afford constant updates can now use AI to maintain “freshness” without high costs.

Who is already experimenting: real-life examples

• Tencent + VISVISE is an example of how a large company is implementing AI tools for 3D art and models, achieving virtually instant prototyping.
• AI NPCs and dialogues – companies like Sony are experimenting with AI engines that allow them to create NPCs that can engage in dialogue, respond to the player, and change their behavior.
• Industry statistics for 2024-2025, according to Unity reports and GDC surveys, more than half (52-62%) of developers already use AI, most often for assets, prototypes, content generation, and back-office tasks.
• AI in live games, AI is actively used for level generation, textures, and optimization, especially in mobile and indie games, where resources are limited and speed is important.

Where AI is not yet perfect: limitations, risks, failures

– Quality is “not immediately AAA.”

AI-generated art and content are good, but often the result needs refinement. Generated images, models, and dialogues can be “flat,” formulaic, lacking the “signature” characteristic of humans.

A bad example is a project where AI generated a background image for a mobile game, and players called it “shovelware” (cheap knockoff), criticizing its quality and graphic monotony.

AI is a tool, not a magic pill. To achieve true depth and quality, competent human control, editing, and design are required.

– Risk of job and professional skill loss

Automating routine tasks means that some specialists may become redundant – especially artists, illustrators, and those who do “monotonous” work. This is a social and economic challenge: retraining, reskilling, and possibly changing specializations are necessary.

– Creativity ≠ algorithm

AI is good at generating content, but it lacks intuition, cultural context, a sense of history, and emotional depth. Games that captivate players with their soul, plot, and characters are still the prerogative of humans.
Yes, it is possible to generate dialogues, NPCs, and quests, but conveying humanism, tragedy, and complex moral dilemmas may be beyond the capabilities of AI alone.

– Legal issues, licenses, data, transparency

When AI learns from huge amounts of data: content, art, music, texts, the question arises: what is legal and what is not? Rights, licenses, compensation, all of this becomes a sore spot. Many in the industry are already warning that without clear regulation of AI content, problems may arise.

– The human factor and quality control

AI cannot simply be “turned on and forgotten”. It needs to be monitored, checked, tested, and edited. Without this, the game may turn out to be a “collection of art,” but not a full-fledged product with soul, structure, and balance.

How to approach AI if you are a developer, producer, or studio

If you are considering whether to incorporate AI into your workflow, here is a step-by-step guide on how to do it effectively:

Assess where AI will truly be beneficial.

• Asset generation (art, textures, models), rapid prototyping.
• Routine tasks: build assembly, testing, localization, optimization.
• Live content elements: random quests, variable NPCs, adaptive dialogues, and scenes.

Launch pilots, don’t scale up right away.

Test AI on specific tasks: for example, generate 10 texture options, 5 NPC options, or a quick build prototype. Compare quality, speed, and costs.

Don’t take away human control.

AI is like a tire changer: it cuts, but the tires are then polished by hand. Everything that AI generates must be edited, design-checked, and tested.

Be prepared for legal and ethical issues.

Make sure that the data, assets, and materials you use are legal. If you are training models on someone else’s art, make sure you have permission.

Invest in competencies.

New roles will emerge: prompt engineers, ML pipeline specialists, and QA designers for AI content. It’s not about “fewer people”, but “different tasks”.

A hybrid approach is key.

The best games will not be “fully AI-generated” but “hybrid”: human design + AI accelerator. This will give you speed + depth + control.

Why future games will be hybrids – not fully AI

• The games that people value are not just about graphics or mechanics; they are about emotions, story, humanity, and cultural context. A neural network alone is not enough to understand what “tragedy”, “love”, “loss”, and “choice” mean.
• AI is good at scaling: more assets, more content, more variety – but the meaning, structure, balance, and soul of the game remain with humans.
• Hybrids allow us to use the best of both worlds: speed and economy where possible; human talent where style, intuition, and artistic taste are important.

What will happen in 3-5 years – forecast

1. AI is a standard tool in game development. Most studios already use AI – by 2027-2028, it will become the norm: asset generation, prototypes, automation.
2. Hybrid games with adaptive, live content. NPCs, dialogues, quests, and the world will adapt to the player; games will become less “linear” and more “organic.”
3. New business models and opportunities for indies. Small studios will be able to make high-quality games with smaller budgets. Competition and creativity will increase.
4. Legal and ethical regulation. Standards for the use of AI assets, licensing, copyright, and transparency will emerge.
5. New professions and redistribution of roles. Artists, writers, and designers will work differently, using AI as part of the pipeline, and tasks will become more creative than routine.

Is it worth betting on AI in games?

If you are a developer, producer, studio manager, or just a game lover who dreams of creating your own, here’s what we think:

Yes, AI is worth using. It is a powerful tool that can save time and money, speed up prototyping, and scale products.

The best games are hybrid games. Those who know how to combine human intuition, style, emotion, and AI efficiency will have an advantage.

AI is not the future “killer” of game development. It is its tool. A good team that knows how to use this tool is the real future of the gaming industry.