Jack Williams, iTero and the AI Grey Zone: When Analytical Edge Becomes an Exclusive Asset

Jack Williams sat down to talk about iTero, Giant X and the future of artificial-intelligence coaching in esports. Two topics were pushed to the front: iTero's exclusive arrangement with Giant X and the likelihood of being copied, alongside the question of whether AI opens a path to competitive cheating. Both sit on the commercial layer and the integrity layer. The space between them — fairness inside a closed league — barely surfaced in the conversation.

Context

Giant X operates in the EMEA region, tied to the League of Legends ecosystem and a franchised, closed-league model. iTero sits on the other side of the table: a vendor of analytics and coaching tools built on machine-learning models. When buyer and seller sit down together, the thing being exchanged is usually called a technology solution. What it actually is, is access.

In football I once built an xG spreadsheet for all 64 matches of the 2026 World Cup and drew one conclusion: tools do not create an edge, people who ask the right question do. My first xG spreadsheet taught me that every goal has a hidden story. Football and esports differ on the surface, but the same data layer sits underneath.

The interview also reveals what it does not supply. No dataset is published, no sample size, no evaluation methodology. That means every performance claim about the tool is unverifiable from the outside. Based on my experience tracking matches, an analytics product without a disclosed methodology should be treated as a hypothesis, not a conclusion.

One historical detail deserves separation. The piece recalls Natus Vincere lifting the Aegis of Champions at Gamescom, 14 years ago. That event took place in 2026, which anchors the article to roughly 2026. It is nostalgic colour from the writer, not a signal about the current competitive landscape; reading it as competitive data would be a category error.

Core analysis

Each title's patch cadence determines the real value of an AI tool, and this is the first variable to go missing. Dota 2 receives infrequent but system-breaking patches from Valve, leaving long stable windows in between; models trained on historical data therefore retain validity longer. League of Legends is patched every two weeks by Riot, sharply shortening the half-life of any learned pattern. The consequence is concrete: in the first title, AI sells model depth; in the second, AI sells speed in detecting the meta delta. Two different products under one name.

With no patch-version information disclosed, the fit between the tool and any specific roster cannot be assessed. This part needs saying plainly: insufficient information, no conclusion possible. Any claim about champion pools or patch adaptation would be speculation.

At the league layer, the issue is clearer. A closed franchised model removes relegation pressure. A permanent member holding exclusive access to an analytics tool keeps that edge across seasons instead of being competed away, as it would be in an open circuit. That is a structural difference, not a difference in talent. A league that permits exclusive tooling has implicitly accepted inequality in preparation.

Publishers do not share one stance on third-party software either. If Valve and Riot hold different standards on data access, the addressable market for an AI tooling vendor differs fundamentally by title. A single product marketed identically across both ecosystems is a flag worth checking.

I don't predict the future by intuition; I only read the traces numbers leave behind. The trace here sits in the publisher's history of intervention. In-game communication was squeezed season by season, from allowing coaches behind players to an outright ban. AI-driven analytics tools are travelling the same trajectory, a few years behind.

Contrarian angle

The AI-assisted cheating debate most likely does not revolve around real-time assistance. Help during a live match is already unambiguously banned in every major title, leaving little to argue about. The genuine grey zone is the between-game window in a BO3 or BO5 — when the team leaves the stage and a model can propose adjustments. If the rulebook speaks only about devices inside the playing room and not about data flowing out of it, the hole is in the data pipe.

The copying worry is also usually misplaced. Copying a product's replica is easy; copying a data moat is not. iTero's value lies in the proprietary dataset the exclusivity deal produces, not in the model architecture. Competitors can rebuild the algorithm in months; they cannot rebuild Giant X's internal head-to-head history.

And the verification condition still applies. No figure in the interview allows confirmation of how much edge the tool produces. That is a serious gap, and it is what I always check before trusting an analytics product.

Takeaway

Every dataset is a scripture, and I am a slow reader. The signal to watch in the next cycle is not an iTero update, but which publisher issues dedicated policy on coaching tools, and whether they force equal access for all league members. When that happens, an exclusive arrangement flips from advantage to risk. For anyone patient enough to wait a season to prove a number.