Hello! Generative AI isn’t just a buzzword – it’s fast becoming a core part of the everyday toolkit for game developers worldwide. This week in AI Gamechangers, we sit down with someone who’s been ahead of the curve. We’re chatting with Tom Pigott, who launched Ludo (before ChatGPT went mainstream) as a platform to help indie developers.

He explains how Ludo’s tools are helping small studios and solo creators harness generative AI to supercharge their design process and shares his perspective on where this technology is heading.

If you’re new here, welcome! AI Gamechangers is a weekly newsletter spotlighting games industry insiders who are putting AI to practical and innovative use. Each edition is a conversation with a leader in the field about the tools they’re building, the problems they’re solving, and their view on where AI in games is going next.

We have some new features planned for the coming months, plus a great lineup of guests already in the works. Got a suggestion, or a project you’d like to share with the AI Gamechangers audience? Drop us a message on Substack or email [email protected]. And maybe we’ll see some of our readers at the Bristol Games Hub’s AI in Game Development session in person on Wednesday.

As ever, scroll down for our curated round-up of AI news and useful tools from around the web.

Tom Pigott, Ludo

Meet Tom Pigott, the co-founder and CEO of Ludo, a platform that provides AI-powered tools for game designers and developers.

After starting in VR and mobile game development, Tom’s team pivoted to building tools that help others create games more efficiently. Ludo launched in 2020 and has evolved to offer several different AI-powered tools that assist with everything from ideation to asset generation. The platform primarily serves indie developers and small studios who want to leverage AI without needing to become AI experts themselves.

Top takeaways from this conversation:

1. AI tech is helping smaller teams to compete more effectively. Many studios plan to integrate AI tools in the next 12 months, with indie and small studios being the most receptive.

2. Ludo's most popular tools include ideation, asset generation, and research capabilities. These tools help developers gain confidence and understand trends without expensive market research.

3. This is not about replacing developers, but augmenting their capabilities. The focus is on handling repetitive tasks and thus enabling developers to focus on storytelling and game design.

AI Gamechangers: What’s the elevator pitch for Ludo, and who uses it?

Tom Pigott: Ludo is the go-to platform for game design for indie developers. And it’s as simple as that. When you have a concept already – or need a new concept – you can go to one place where you have all the tools available to begin the process of designing your game.

We have eight different tools, all AI-powered tools that help you create better, faster and cheaper games. We’re primarily for mobile, but we also [support] PC-based games.

Our clients are mostly small indie studios, solo devs, other creatives, modders. We certainly have some medium-sized studios who are clients as well, but I think there’s a greater willingness to try new types of tools and ideas as a small studio or solo dev because you’re trying to figure out how to get the whole thing done. You’re wearing a lot of hats, whereas in larger studios the workflow is fairly well defined, so getting integrated into that workflow is more of a challenge. I think that’s why indie studios are so receptive to trying out the new tools as they come.

Please tell us about your journey to this point. You launched Ludo even before ChatGPT became mainstream?

Like a lot of studios, we were a struggling mobile game company. Before that, we were in VR, actually. I met my co-founder early on in the process, and he helped us understand some of the opportunities in AI.

As we all know, it’s extremely difficult not only to make your first hit game but also to make multiple hit games. It’s like being in a rock band, trying to get that first and second hit song. The light-bulb moment was when we realised we could have a longer-term opportunity by building the instruments – by building the technology to help other studios, too.

They always say there’s a little bit of a magic sauce for people who know how to make hit games, right? You can have all the components, but then there’s just “something”, especially once you do more than one. We realised that was not our forte.

We could see where the AI models were going; the LLMs were only going to improve and become more open-source. And you could see the struggles and challenges and the pain points that all game developers face. So my bet was this: if we could somehow marry those things and make those AI tools available, there could be an interesting opportunity.

This was pre-GPT-2. From the start, there was a misperception that maybe you could just push a button, and an AI would generate a hit game for you! There was a lot of working with [PR agency Dimoso], educating people, backing up and saying, “No, the models are not magic.”

The most realistic opportunity for us, a small company, was in the ideation possibility. That is still one of our best-selling tools.

I spoke with some game studios, and they said, “It’s not about the idea; it’s about the execution. Ideas are a dime a dozen. If I come up with 1000 ideas, we still have to make a hit game that has merit.”

But what we found is actually coming up with original ideas, new twists and new concepts, is difficult without assistance. You’re just brainstorming, putting the yellow stickies on a whiteboard. And so I think people realised they can create more fully developed concepts with AI tools. That’s where we started.

What are the other most popular tools in the Ludo suite?

The asset generator, which is relatively new. It’s not surprising people want to be able to create usable assets that they can plug into Unity. Image generation, too.

Also, the market research capability. One of the things that we thought was going to be helpful is providing studios with a little bit of market knowledge – not just about the top charts, but also the trends. What are the mechanics that are popular? What are the different geographic regions that are doing well in this type of game? Services like Sensor Tower are pretty pricey for a small studio. We pack all that in; it’s not quite as detailed or robust, you’re not diving into monetisation, but it’s been extremely well utilised. We have a Ludo score, which is an aggregate of four different types of trends. It gives a degree of confidence: “Should I go forward with this concept?” For small studios, that’s a time saver.

Did you receive any investment and funding for the product?

It was fully self-funded. That takes a bit of the pressure off. It takes time for this market to grow, and the AI models to get you know better and better. We’re fortunate that we have a longer-term perspective and are able to grow this out the right way.

There are no guarantees with any of this, but we’re self-funded, and I wouldn’t rule out, ultimately, finding the right partner if that makes sense, if we’re aligned; it doesn’t even necessarily have to be venture capital, it could be an industry partner.

But for now, we are heads down, just continuing to improve it every week, getting more and more well-known in the industry.

Are you seeing more acceptance of AI in the workflow now than you did a few years ago?

It’s gone from scepticism and fear (“Is this going to take my job?”) to much more acceptance.

It’s like when the personal computer came – I’m sure some people lost their jobs. The internet was similar. But you adapt. AI is a technology at the end of the day; you still have to create a good game and execute on it and monetise it.

The reports and surveys out there show that 75% of all studios plan to integrate some type of AI tools in the next 12 months. If they haven’t already. So it’s no longer about going to those early conferences and trying to convince people this is coming, and it’s only getting better.

That’s the amazing thing about it. I’m so excited to still be building this company. We joke that if we get stuck on some problem, we just wait three weeks, and a new open-source model will come out that we can integrate! That’s how fast-paced it is.

One big learning is that almost all studios have told us, “We don’t want to be the AI experts. We don’t want to hire machine learning specialists. We just want to design a game. So we need help, and we just want a solution.”

As you say, the AI scene moves so fast. A new model, like DeepSeek, could appear any minute! How do you stay on top of everything?

My co-founder has a PhD in AI. Jorge [Carvalho Gomes, CTO] is based out of Lisbon. Jorge tests out all open-source models. (If it’s a closed model or not commercially available, then we have to wait because we’re a small team.) He’s on all Reddit subreddits and all chat forums; he’s got a pulse on what’s new.

We continuously test what makes sense for us. We keep broadening our platform and integrating all of them, and I think that has been one of the things that’s kept us around. We’re not just image generation and just ideation. We’ve continued down the tracks, heading towards game development itself.

Ironically, one of the benefits of having so many smaller customers is that one big customer is not going to knock you on your heels if they go with somebody else. It’s very much a long-tail, long-term approach of getting 1000s of small solo devs, small studios, and modders.

Why do people create games? I think that’s the interesting aspect. It’s not necessarily that this is going to be their profession, their sole job. They might be building with their friends.

UGC is a huge trend. Is Ludo a good tool for people who want to build for, say, Roblox?

Correct, and other platforms like that, and even some of the social media games and web3 games. They don’t want to go through the hassle of necessarily applying and going on to the app store – but they want creative expression. It’s an art form of sorts.

There’s more openness around AI now, and people are willing to try the tools. But there are also questions around ethics. How do you see AI tools like yours fitting into a landscape where people are worried about their own jobs, and about human creativity in the industry?

I don’t think you can pretend it won’t affect certain types of roles. If your job is a stock photographer, right? Or a basic 3D artist?

There are very few industries, really, that aren’t thinking and talking about AI. Maybe coal mining or something.

It feels like there are so many models. You’re not going to put it back in the bag, so to speak. Whatever legislation or regulation there may be, the impact is here to stay, across all industries, and gaming is no exception.

So, will it affect some jobs? Yeah, but I’m on the side that AI tools models will ultimately create more, different jobs than there are now. People get worried about automation (“All I have to do is push the button, and the whole thing is done!”). But it’s the creative aspects that are tough.

And what kind of games will people be able to make? Will there be some new types of original AI games? What does that even look like? There are a lot of companies thinking about it. I think there could be whole new experiences in the next five years. We can all hang in there: five years, good things will probably happen.

Do you think changes in legislation around data, copyright, training and so on are going to impact the AI industry?

We don’t really think that it’ll have a massive impact. The privacy issues themselves are so ubiquitous – it’s a challenging, challenging area. I don’t think that there’s going to be something that will effectively make it so that only your own data is the one that can be utilised or siloed. Open source is really the trend in terms of where things are.

We get questions from the perspective of a developer. “Hey, is Ludo using my brainstorming ideation? Are we helping to train your models?” And we say, “Absolutely not!” Or, “Does Ludo own my creative output?” And we say, “Absolutely not!” You own whatever you create on the Ludo platform, 100%. If you can make a million dollars off that game, you keep a million dollars. We’re not taking anything and that gives reassurance. People are still trying to understand the rules and regs for AI usage.

What sort of feedback have you received from the studios you’re already working with?

It’s been great. Remember, the majority are 1-10 persons, so they’re probably companies you haven’t heard of.

But the feedback has been terrific in terms of use cases for helping them design games with more confidence. They also have suggestions for future tools and technical product roadmaps they’d like to see. Of course, everyone wants the AI to do it all for you! But realistically, the big things coming are video, prototyping, and text-to-video. If you can do that, you can begin to rapidly test basic ads. So then, all of a sudden, you don’t have to rely on large publishers as much.

Gaze into your crystal ball – how much of an impact is AI going to have on games in the year ahead?

From a game developer’s perspective? It’s not something that you can just ignore or hope goes away! You have to adopt and adapt. Try it out. Test it. Test us, test our competitors. Test tools and get smart about where AI fits in your workflow. If you’re a small or medium-sized developer, you’re wearing a lot of hats, so maybe it’s a specific need you have for playtesting, maybe QA. Maybe it’s other things.

Broadly, the AI models are getting better. The amount of research papers being released is going up exponentially. It shows you there’s still more to be done. The people who are creating the AI models have not maxed out on what the capabilities are. You’re just scratching the surface of where things could go, for both good and bad. There’s going to be some disruption. There’ll be some misuse, bad actors, and all of that.

But for the next year as a game developer, no matter what your role is, if you’re not already using or testing, I would highly recommend beginning to understand how AI tools will impact your role.

The genie is out of the bottle now. The world just saw Nobel Prizes handed to AI companies. AI is out there in the world for people to see what it can do. What else does the future hold?

A lot of people have said, “What types of companies will be built in the future?” Sam Altman said it’s perfectly feasible to think that a two-person company could create a billion-dollar game with AI tools.

I think the structure of companies, the size of companies, will change. There’s traditionally been a need to scale up – get as many programmers on board as you could, and that gives you your moat. You’re hiring all the best talent. Those days are changing! Obviously, the biggest companies will continue to have huge workforces, but I think for smaller companies, it’s never been a better time to be an entrepreneur. You have so many tools. Things are inexpensive. All the SaaS options. You can get going much easier than you could five years ago.

What are your plans for Ludo now?

We just keep our heads down and keep trying to improve on a week-by-week basis, adding more value for our users, with more product and better versions of our existing products.

We’re big believers in the growing markets: Brazil, MENA, India, Pakistan. And trying to better understand how we can have more of a presence there. So I think for us it’s a growth period. We’ve got a great product-market fit. We’re still pretty small and under the radar, but now it’s time to get out there and talk to lots of the world.

I believe in small giants. You can build a great company where cash flows wonderfully. It doesn’t have to be a billion-dollar company, right? Unfortunately, some of the processes of a VC-backed company make you think if you’re not a billion-dollar unicorn, then you sort of failed. And I think there’s so much opportunity for a small, extremely profitable company to just keep going.

Further down the rabbit hole

What’s been happening in AI and games? Here’s your essential news round-up:

• The UK’s Bristol Games Hub is hosting a session called “AI in Game Development - Hype, Hope, Help or Hell?” on the evening of Wednesday, 23rd April.

• There are just a couple of weeks until the Practical AI track hits the Dubai GameExpo Summit. Tickets are available from Platinum List.

• Gemini got further than Claude in Pokémon. But was it a fair test?

• Tamay Besiroglu’s startup Mechanize has come under fire for its stated mission to replace all human labour with AI.

• VentureBeat’s Lavanya Gupta sums up why it’s not as easy as you think to switch LLM models in your software.

• Open source AI outfit Eliza Labs has launched auto.fun, a no-code platform enabling users to create autonomous AI agents that can perform complex blockchain actions.

• Decentralised AI platform 404-GEN has released the first blockchain-based text-to-3D model generation plugin on Unity's Asset Store.

• Saga’s Rebecca Liao spoke about turning Titanus, from the mobile game Hero Hunters, into an AI agent. You will hear more from Rebecca in a future edition of AI Gamechangers.

• Turkish startup Boby.ai has raised $1.25 million in seed funding to develop AI-powered mobile apps. It was co-founded by Gökçe Nur Oğuz, who previously led Playable Factory and Fabrika Games.

• NVIDIA plans to manufacture American-made AI supercomputers in the US for the first time, in Arizona.

• Games industry veteran Chris Heatherly vibe coded a satirical “meme game”.