"The end of SaaS"? Why AI will not replace real igaming platforms

In December 2024, Microsoft CEO Satya Nadella declared on the BG2 podcast that “SaaS is dead”. That statement sent a shockwave through the global technology industry. Many were provoked by the idea, especially since software-as-a-service has shaped enterprise computing for nearly two decades. Against this backdrop, the question is especially relevant for highly regulated industries such as igaming, where software is only one part of a much larger operational architecture.

Here, Denis Kosinsky, COO at NuxGame, gives his thoughts on why AI cannot replace the platform discipline required in regulated igaming.

EGR: The “SaaS is dead” narrative is getting louder as AI agents promise to build and run software for businesses. In igaming specifically, what does AI genuinely enable today, and where do you think the hype is overstating what operators can safely rely on?

Denis Kosinsky (DK): The “SaaS is dead” headline is catchy, but SaaS still represented over 10% of IT spending in 2024 and remains the foundation of enterprise software worldwide. What AI is really changing is how people interact with that foundation: fewer dashboards, more automated workflows and a shift toward execution rather than logins.

That said, in igaming, building software is not the hard part – operating it safely is. A real igaming platform handles regulatory logic, auditability, risk controls, payment integrity and uptime under real-money load. That is why automation alone is never enough when a regulator requests a full transaction trail, or a KYC escalation happens mid-session. The system must respond consistently and in line with defined rules.

AI can accelerate development, but it does not carry responsibility in a licensed environment. Think of AI as a fast drafting tool, while fully governed platforms are built for production reality, where regulators, players and financial systems all meet.

EGR: If AI builds software fast, does that make platforms interchangeable? From your experience, what truly separates a software product from a real igaming platform?

DK: Many people now assume that if AI can generate software quickly, platforms become interchangeable. From an operational standpoint, that simply is not how regulated businesses work. When real money is flowing, uptime is revenue, and player protection is a regulatory obligation, the platform is not solely a technical choice; it is a long-term business decision.

And that is exactly where the difference shows up. You can replace a feature. But you cannot easily replace the embedded compliance logic, fraud controls, reporting standards and live-market stability that a mature platform accumulates over years. Operators learn very quickly that ‘working software’ and ‘licensed operations’ are not the same thing.

In live operations, the truth comes out fast. The real platform test is a Saturday night surge, a payment provider delay and a compliance report due by Monday morning – all at once. That is why solid platforms remain distinct – and why experienced partners are built around resilience, governance and trust (as opposed to interchangeable code).

EGR: Licensing authorities increasingly demand auditability and clear accountability. Where do the biggest compliance risks appear if an operator relies on AI-managed systems without an experienced platform partner?

DK: What we see in practice is that AI is pushing software from ‘screens and seats’ toward automated outcomes. That shift is real, but in igaming, outcomes carry legal exposure. The most serious regulatory risks occur when AI-driven systems function without compliance controls built in. Regulators require audit trails, verifiable payment handling, explainable risk logic and consistent enforcement across jurisdictions. And enforcement activity in regulated markets is only increasing.

A good rule of thumb is this: AI can do the work, but it cannot take the call when the regulator asks, “Who approved this?” That is where mature igaming ecosystems like NuxGame add real value. We are built to take that call – promptly and professionally – because in regulated igaming, someone always has to own the answer.

EGR: Looking forward, if AI keeps accelerating development and automation, what will operators still need from established platform providers over the next 12-24 months – specifically, in terms of trust, regulatory responsibility and lasting stability?

DK: What matters for operators is that AI sits on top of existing systems to automate workflows – it does not replace governance or infrastructure. In regulated igaming, that distinction defines survival.

In the coming 12-24 months, operators will still rely on established platforms for one reason: AI does not remove operational risk, it shifts that risk into new layers of automation, monitoring and control. Trust will come from systems with proven uptime, controlled payment infrastructure and compliance frameworks that stand up in audits. Platforms also carry institutional memory – years of payment exceptions, fraud patterns, regulatory nuances and real incident responses that automation has never ‘lived through’.

The operators that succeed will not be the ones chasing the fastest AI-generated build; they will be the ones running stable, auditable businesses when scrutiny increases. AI is a fast sports car, but regulated igaming still needs brakes, seatbelts and a licensed driver. In other words: AI may reshape SaaS, but replacing top-level igaming platforms at scale – not realistic.

Denis Kosinsky

Denis Kosinsky is the COO at NuxGame. He oversees platform strategy, product development, and advanced technology solutions. Kosinsky works with AI features, gamification tools and Web3 integrations to help operators launch and grow strong igaming businesses worldwide.