In an industry where margins are thin and player loyalty is fleeting, customer experience has become a key differentiator for operators. As AI becomes a core operational requirement, leadership teams face a clear choice: build proprietary technology in house, or partner with purpose built AI CX providers.

Alex Gould, CTO at Conduet, explains why more operators are choosing the latter.

What industry-specific CX challenges can an exterior solution address ‘out of the box’ compared to a generic build?

Generic AI struggles in sports betting and iGaming because player inquiries are shaped by complex, domain-specific rules and edge cases. Questions about settlements, promotions, withdrawals, or cash outs are rarely straightforward. They depend on wager structure, timing, eligibility criteria, and operator-specific logic.

Over 80% of player inquiries require pulling live, account-specific information from the PAM and applying it correctly within that broader rule set. Without purpose-built logic to interpret both the data and the edge cases around it, responses quickly become incomplete or incorrect.

This limitation is reflected more broadly in enterprise AI adoption. Research from MIT found that 95% of enterprise AI initiatives fail to deliver measurable business impact, often because broadly trained models are pushed into live environments without the domain context needed to handle real-world variability. What appears to work in controlled testing breaks down once exposed to operational complexity.

Purpose-built platforms are designed around this reality. By training on gaming-specific data, workflows, and failure modes, they can interpret live PAM data in context and handle both common and complex inquiries accurately from day one, without relying on extensive rules, manual escalation, or post-deployment patchwork.

How would you characterise the current skills gap within operator teams regarding AI implementation?

Operator CX teams are closest to the customer and understand where friction exists. The challenge is not identifying opportunities, but delivering AI that performs reliably in production. Turning insight into production-ready capability requires technical depth, dedicated ownership, and sustained iteration that sit outside the remit of most CX organisations.

Deploying AI in gaming requires expertise across model evaluation, conversation design, failure handling, and real-time interaction with PAMs and ticketing systems. It also requires ongoing investment to monitor performance, manage edge cases, and improve outcomes as volumes and player behaviour change. CX teams are structured to run day-to-day operations, which makes sustaining this work in parallel difficult.

As a result, many internal AI CX efforts stall or remain narrow in scope, not because the opportunity is unclear, but because the execution burden is too high.

What is the average time to market using a specialist platform, versus a full in-house build?

In-house AI efforts typically take 18 to 36 months to reach enterprise-ready scale. The delay is driven by the need to coordinate across CX, product, data, and engineering while establishing new ownership and operating models inside live CX environments.

A specialist platform compresses this timeline materially. With gameLM, operators can move from concept to live inbound CX in six to 12 weeks. Operators achieve 60%+ resolution within 90 days, scaling toward 80%+ shortly thereafter.

Why does a purpose built partnership model matter in iGaming & OSB CX?

In iGaming and online sports betting, the challenge is not adopting AI, but making it work reliably at scale. Generic platforms often shift the burden onto operators after deployment, requiring significant time and internal effort to adapt the technology to gaming-specific realities. That effort compounds as complexity grows.

A purpose built partnership model changes that dynamic. Instead of operators spending months closing gaps, AI is deployed using operating patterns already proven in live gaming CX. Common failure modes, escalation paths, and performance tradeoffs are understood upfront, reducing the need for downstream rework and ongoing firefighting.

Conduet applies this approach through gameLM, informed by operating a 500+ agent gaming CX organisation. That operating knowledge functions as an embedded R&D capability, shaping how the platform is tuned, prioritised, and extended alongside each operator’s environment. Inbound CX performance today directly informs the development of additional, gaming-specific capabilities such as reactivation, payments optimisation, and fraud prevention.

The result is a partnership model that delivers strong outcomes without transferring the hidden cost of adaptation and maintenance back to the operator, allowing CX capability to keep pace as the industry evolves.

Alex Gould is the CTO at Conduet, where he leverages his technical and strategic background to guide technology strategy and innovation. He is also the Founder and CTO of Everyday AI and previously founded computer vision company ViewX. Alex’s earlier experience includes roles at Primary Venture Partners and Bain & Company, and he holds an MBA from Columbia Business School and a Bachelor of Engineering (Hons) from the University of Canterbury.

The post Why operators are choosing to buy in their AI strategy appeared first on Eastern European Gaming | Global iGaming & Tech Intelligence Hub.