Beyond the Buzzwords: The Real Impact of AI
AI has become one of the most frequently cited - and least clearly defined - concepts in the betting industry
From game development and personalisation to operational efficiency and player insights, AI is now positioned as a transformative force across the supplier ecosystem.
Yet as its use accelerates, so too does the challenge of separating genuine innovation from marketing shorthand. With the help of iGaming experts, G3 examines how AI is being applied in practice today, where it is delivering tangible value for operators, and where expectations still exceed reality.
Rather than focusing on theoretical potential, our roundtable discussion centres on live, deployable products and services, the commercial and regulatory realities shaping adoption, and the responsibilities suppliers carry when communicating AI capabilities to the market.
As operators become more discerning in how they assess AI-led solutions, the conversation also looks ahead, identifying which applications are likely to become business-critical over the next 12 to 24 months, and what success will realistically look like for AI in casino gaming beyond the buzzwords.
AI is now embedded in much of the industry’s marketing language. In practical terms, what constitutes a genuine AI application in casino games today, and where is the term being overstretched?
Andreas Koeberl, CEO at BetGames: The industry is incredibly behind. ICE was a good example. You read “AI” quite a lot, but there wasn’t anything radically new. I was surprised that our Casino Designer product was one of the few key highlights in that space. There are some nuances of AI in the games development process, but when you look at what happens in the gaming space outside iGaming, we are laggards.
Sam Depoortere, Chief Product Officer at OpenBet: AI in sports betting has evolved from theoretical concepts to real operational tools enhancing player experience, product development, and scalability. However, the term “AI” is often applied loosely, with many traditional machine learning models rebranded as something more advanced. True AI involves systems capable of reasoning, autonomous task execution, and natural language understanding – particularly agentic frameworks and large language models.
At OpenBet, this includes generative responsible gaming assistants and workflow automation solutions that go beyond simple predictive modelling. These tools allow deeper contextual insights, faster analysis, and more intelligent decision support, marking a clear shift from legacy ML toward genuinely transformative AI applications.
Bjørnar Heggernes, Chief Commercial Officer at The Mill Adventure: For me, at least when it comes to casino games, AI is most useful when it removes repetitive or procedural work so teams can spend more time on the things that actually matter – creativity, analysis and decision-making.
On the analytical side, AI can help teams spend more time understanding the numbers and what the data is actually telling them. Operators track hundreds of metrics to measure performance, and it is easy to get stuck focusing only on active players, wagering or revenue. The more interesting question is what hidden-in-plain-sight insights AI could help contextualise and present.
Where AI is being overstretched today is in the design and delivery of games. Very few operators need more content or faster content, and even if AI could deliver branded games more quickly, that does not mean players will prefer them. Marketing and branding help, but they do not automatically translate into growth.
What are the most meaningful AI-driven products or services suppliers are delivering right now – solutions that operators can actively deploy rather than concepts still in development?
Andreas Koeberl: Today’s most deployable AI solutions sit firmly in the backend, powering CRM, player segmentation, analytics, and automated decision making. These tools help operators personalise engagement, optimise retention strategies, and streamline workflows by analysing behavioural patterns at scale. While technically advanced front-end ideas – like AI-driven presenters – exist, real-world tests show they have limited commercial impact.
For now, the highest value innovations are those improving operational efficiency, accelerating development cycles, and enabling teams to iterate faster without adding headcount. So, we mainly talk about margin expansion here. These solutions deliver measurable benefits immediately, while more experimental, player-facing AI applications remain interesting but commercially unproven.
Sam Depoortere: Operators can already use high-impact AI tools that move beyond experimentation. OpenBet’s Neccton solution provides real-time player protection capabilities by analysing behavioural data with machine learning to identify risk indicators. The newest version incorporates a generative AI-enhanced assistant that deepens player analysis and provides more actionable insights for support teams.
AI is also advancing trading and BetBuilder accuracy, strengthening pricing models and risk assessment. Beyond customer-facing features, AI now streamlines internal processes such as metric tracking, prototyping, and product lifecycle management. These operational enhancements shorten development cycles and help operators deploy better products and protections more efficiently.
Bjørnar Heggernes: I will start with the obvious, and I am sure even some of the contributions on this panel have likely been refined with AI, but copy and content writing, as well as general content optimisation stands out as a big one.
Tools such as ChatGPT, Claude and Copilot allow teams to prompt ideas, structure frameworks and generate more relevant content much faster than before. Better still, no one on LinkedIn reached out asking for a quick chat to sell it to me.
They have been organically adopted and become part of everyday use. In many ways, this is already the most widely used AI product in the industry today, and companies that have not adopted it in some form are likely already trailing. That said, proper training is essential to ensure these tools are used correctly.
Beyond that, we are also seeing meaningful progress in areas such as AI-driven customer support chat tools, BI tools that allow instant creation of reports and graphs through prompts, and AI-assisted tools that can quickly generate websites, wireframes and product concepts. None of these should replace people, but they can significantly improve how teams work.
Where is AI delivering measurable benefits for operators today, whether in efficiency, scalability, player engagement or content optimisation, and how should that value be assessed?
Bjørnar Heggernes: The most concrete example I have is our SmartLobbies solution, which has allowed operators on our platform to fully automate casino lobby management for the past six years while providing players with personalised game recommendations (think Netflix!).
Instead of manually positioning games every day to accommodate new releases and providers, the system dynamically promotes relevant content based on player behaviour and performance data. This removes a large operational burden for casino teams while also helping surface games that players are more likely to engage with.
In practice, this translates into longer session times, better visibility for relevant games and reduced operational effort managing lobby content. From an operator perspective the value is clear – improved player engagement, lower operational cost and the ability to scale content management without adding additional resources.
Sam Depoortere: AI is producing tangible improvements across efficiency, scalability, engagement, and responsible gaming. Automation reduces manual reporting and routine analytical work, enabling specialists to focus on strategic tasks. Scalability is supported by shared AI APIs that unify capabilities across trading, risk, and compliance functions. Neccton strengthens player safety by identifying harm patterns earlier, improving regulatory outcomes and long term retention.
Operators also see gains in innovation speed, margin protection, and more precise risk management processes. These benefits become meaningful when measured through standardised KPIs, ensuring investment in AI translates into demonstrable performance improvements rather than theoretical value.
Andreas Koeberl: AI delivers the clearest value in efficiency and operational optimisation. Machine learning systems enhance analytics, enabling faster insights, predictive modelling, and improved decision-making.
Automation reduces manual effort across production, marketing, and back office tasks, allowing teams to scale output without increasing resources. These gains translate into better resource allocation and faster iteration cycles. This is more evolutionary than disruptive.
As with any new technology, the starting point is a clear use case, a pain point you want the technology to address. Start small and expand quickly. Most AI projects fail to deliver AI, because companies see fancy demos on social media and thing AI is magic.
It’s not. If you have an unclear and fuzzy scope, you will not succeed, e.g. if you look at game content production start with games assets or sound, then expand. Look at what consumes most time and becomes a bottleneck, then try and solve it with AI to unlock immediate value.
As AI enables greater automation and personalisation, how are suppliers ensuring transparency, regulatory compliance and appropriate human oversight remain firmly in place?
Bjørnar Heggernes: Any AI tool must be grounded in regulatory compliance. If we move slightly away from the traditional discussion around leveraging AI for CRM personalisation and instead look at how it can support responsible gaming, risk monitoring and fraud detection, the value becomes even clearer.
Rules, thresholds and limits must act as guardrails and be embedded into any personalisation or automation framework. They create the anchor points for intervention, whether that means triggering alerts for monitoring, flagging risk behaviour or initiating automated communication with players.
Beyond those points, AI can help operators detect patterns that would otherwise be difficult to identify at scale. However, the key principle remains that these systems support human decision-making rather than replace it. Clear escalation processes, manual review of high-risk cases and transparent reporting ensure that automation operates within defined regulatory and ethical boundaries.
Sam Depoortere: Responsible AI deployment requires transparency, strong governance, and human oversight. Suppliers are increasingly pairing advanced cloud and AI tools with robust responsible gaming technology to ensure players and operators are protected.
Crucially, AI is positioned to enhance – not replace – human judgment, with teams remaining accountable for key decisions. Governance structures are becoming more formalised, with dedicated roles and cross functional ethics committees reviewing model logic, compliance, and operational risk.
OpenBet’s own multi disciplinary AI ethics committee ensures all insights are explainable, auditable, and aligned with global regulatory expectations. This framework ensures innovation progresses without compromising safety or trust.
Andreas Koeberl: Like with any technology, you must understand it first and have an overview of your dataflows and your guardrails in place. You can host almost any model in the EU already and can get ISO certified for AI native applications. There isn’t any magic behind this.
Of course, these models are a black box, and you need to be careful how much autonomy you give them, but at the end of the day, the human remains in control. At least for now.
Has the industry been guilty of over-promising what AI can deliver in the short term, and how should suppliers better align AI narratives with operational reality?
Sam Depoortere: Across many industries, AI has been positioned as a quick plug-and-play solution, sometimes leading to unrealistic expectations. In reality, meaningful impact depends on strong data foundations, responsible governance, cultural readiness, and long-term commitment. OpenBet approaches AI as an ongoing transformation rather than a rapid fix, prioritising infrastructure and operational discipline.
Its multi-cloud strategy – working with providers like Anthropic, Google, Microsoft, and Amazon – ensures flexibility in a fast-moving space where new capabilities can emerge daily. This mindset allows continual re-evaluation and prevents operators from being locked into any single pathway. Sustainable AI success relies on iteration and adaptability.
Andreas Koeberl: I haven’t heard many promises to be fair. People talk about AI in vague terms, but not many have presented AI native products, meaning tools built from scratch with large language models (LLMs) and machine learning at their core. So, from that perspective – no.
Bjørnar Heggernes: I would not necessarily say the industry has over-promised. AI is clearly here to stay and is not another hype train that will simply pass by. What we are seeing instead is that most companies are still learning how to adopt it properly and implement it safely within their organisations.
As with any emerging technology, there will always be the select few trying to sell quick solutions for problems that may not really exist. AI was never meant to be a silver bullet. It is simply another tool, and for companies that have adopted it thoughtfully, the benefits are already quite obvious.
Which AI capabilities are most likely to shift from competitive differentiators to baseline expectations for casino suppliers over the next 12–24 months, and what will truly set suppliers apart once that happens?
Andreas Koeberl: There are two aspects: Firstly, margin gains through automation along the game production chain. And secondly, actual long-term competitive gains by leveraging AI to produce new genres that are not possible without AI. The latter will be the disruptors. The ones that get acquired by the big hitters in the iGaming space.
Sam Depoortere: Within the next 12 to 24 months, AI-driven player protection and automated trading adjustments will shift from being high-end differentiators to baseline expectations, and this is the primary reason we have invested in the right tools, platforms and people to ensure we have the right foundations in place.
Suppliers who do not offer these will quickly find themselves uncompetitive. Once these features become standard, what will then set a supplier apart is the level of integration and the scale of their AI ecosystem.
Offering a standalone AI tool is a start, but it is nowhere near as effective as providing an integrated solution where trading, risk management, geolocation, and product development all feed into a singular, intelligent engine. We are focused on moving beyond individual projects to create a consistent, scalable impact will lead the industry in terms of responsible, compliant, impactful innovation.
Market leaders will embed AI as a core operating layer, driving better experiences, faster execution and smarter decisions.
Bjørnar Heggernes: Beyond gaming content, the broader supplier ecosystem will increasingly be expected to ensure data is increasingly more consumable, whether that be via prompts or generative summaries that provide key insights and measurable action points.
If I could compare it to fitness and health trackers, WHOOP has revolutionised how data is turned into meaningful output and simplified action points. Not getting enough sleep? WHOOP will tell you exactly why, how to fix it and why you should do it, based on your personal health metrics. Beyond that, it has built a vast knowledge base and articles based on extensive research, all available at your fingertips through a simple prompt.
As these kinds of AI capabilities become increasingly standard across the iGaming industry, the real differentiator for casino suppliers will be how effectively they translate complex data into clear, personalised insights that empower operators to make faster, smarter decisions.
