Jack Williams, iTero and the Between-Game Window: The Unwritten Boundary of AI Coaching in Esports

Introduction

In an interview centered on iTero — an artificial-intelligence coaching tool — Jack Williams spends most of his time on two subjects: the exclusive agreement with GIANTX, and the likelihood that his product will be copied. Not a single data table appears. No test sample, no win rate is offered for verification. The crowd will overlook that. I will not: in esports, a data gap is usually where the real story is hidden best.

The only number that can be anchored does not come from the interview itself but from the writer's biography: Natus Vincere lifted the Aegis of Champions at Gamescom "14 years ago." The International 2026 took place at Gamescom in 2026. The subtraction yields 2026. This conversation was recorded at a moment when AI had already slipped into the strategy room, but the rules of the game had not yet been written.

The match has stopped, but the data stream keeps flowing forward. Jack Williams' question is not technical. It is a governance question.

Context

To read this interview correctly, it must be placed in the context it belongs to.

In-match assistance — real-time assistance — has been clearly banned in every major title for years. No publisher permits software to whisper into a player's ear while they are laning or fighting. So when people argue about "AI coaching," what is actually on the table is not the in-match tool. It is the pre-match, between-game, and post-match tool.

Of those three windows, the between-game span is the most sensitive. A BO3 or BO5 has a short window — often just minutes — when the coach is allowed to walk in, close the door, and say anything. No one inspects the content of that conversation. No one knows whether the coach is reading from memory, from a sheet of paper, or from a model that has just computed the opposing team's win probability when they play toward the bottom lane.

That is the real grey zone. Not the tool itself, but its timing and its channel of information delivery.

On the organizational side, GIANTX is the central name. It is reported to be an organization operating within the League of Legends ecosystem in the EMEA region, formed from the merger of two well-known organizations. If so, the governance framework applying to the iTero arrangement is Riot Games' set of rules on third-party software and competitive integrity. With Dota 2 — a title mentioned only as a memory — the governing party is Valve, and Valve's treatment of third-party tools has historically not matched Riot's.

That difference is not a small detail. It is a first-order commercial variable, and it is entirely absent from what the public gets to hear.

Core

This is where I want to pause longer, because most readers will see the headline "AI coaching" and immediately think of whether a machine can replace a coach. That question is wrong. The right question is: how long does a machine-learning tool survive before the meta shifts beneath its feet?

The answer depends on each title's patch cadence, and this variable is one the interview never touches.

League of Legends runs on a two-week update cycle. Each update brings champion stat changes, item changes, and sometimes shifts in match tempo. For an AI model trained on historical data, that cadence means the lifecycle of a learned pattern is compressed very quickly. The value of AI here is no longer "solving the meta," but "detecting the meta shift faster than the opponent." That is an advantage in speed, not in knowledge.

Dota 2 moves on a different rhythm. Valve ships fewer major versions, but each change is systemic and structure-breaking. Between such changes lie long stable windows. Within those windows, a machine-learning model based on historical data retains value longer. Here, AI is rewarded for the depth of its historical modeling, not for its speed.

This leads to a conclusion few state aloud: if one AI product is marketed identically for both titles, that is a warning sign. Its value inverts between the two environments. A tool optimized for Dota 2 may be mediocre in League of Legends, and vice versa.

GIANTX, if it is indeed a member of a franchised league such as the LEC, makes the competitive context of the exclusive agreement even more important. In an open circuit, weak teams can be relegated, strong teams can be replaced, and any structural advantage faces pressure to be competed away over time. In a franchise league, every member is permanent. There is no relegation. A structural advantage held by one member persists across seasons instead of being erased by competition.

That is why an exclusive tooling agreement carries more structural weight in a franchised league. It is not merely a commercial matter between one company and one organization. It is an implicit decision by the league that inequality in preparation is permitted to exist.

I do not believe in the hand of fate; I believe in the data curve. And that curve says that as competitive rewards grow, pressure on the league operator to standardize tooling access will grow too. This is the path already walked by in-match coach communication rights: first a grey zone, then progressively tightened into explicit regulation.

Based on my experience following knockout matches across various tournaments, I have found that the between-game window is where matches are often decided. Many BO5 series are decided not by game one, but by the adjustment in game three. If a tool can compress that adjustment from three minutes of collective thinking into ten seconds of lookup, the advantage is not small. It does not win the game for you. It buys you thinking time — the scarcest resource in a tense series.

On the risk of being copied, Jack Williams says plainly that it is possible. I read this detail as a commercial signal, not a confession. In the analytics-tool industry, competitive advantage does not come from the algorithm — because algorithms can be copied. It comes from data access, from exclusive relationships with organizations, and from speed to market. The exclusive agreement with GIANTX is the protective barrier, not the product.

If the product is easy to copy, the real asset is the contract. That is the logic of every commercialization strategist.

Contrarian Angle

The question the interview does not ask is the question of competitive fairness. The two revealed sections — working exclusively with GIANTX and the likelihood of being copied, alongside AI-assisted cheating — revolve around the commercial frame and the integrity frame. The frame between them is left empty: whether an exclusive tool creates an uneven playing field within a closed league.

The crowd sleeps within emotion; I stay awake with the data. The issue is not whether iTero is legal. The issue is that if it is legal for one team and not for the other nineteen, then "legal" becomes a meaningless term. Competitive integrity is not only about fighting cheating. It is also about ensuring that advantage comes from capability, not from access.

And here I must be careful with myself. My side profession is betting analysis; I have an incentive to read every number in a direction favorable to some wager. So I separate them: the data findings above stand independently of the recommendation below. Whether iTero actually produces a measurable advantage is something I cannot confirm, because the public record contains no test sample, no data size, no evaluation methodology.

The biggest mistake is not placing a bet, but placing it with the crowd. And the crowd right now assumes that "AI coaching" is a revolution. I assume it is an unresolved governance problem — one capable of shifting the commercial landscape of an entire league before the rules are written.

Takeaway

What is worth watching next season is not which tool is more accurate. It is which publisher will be the first to put pen to paper on a regulation governing access to analytics tools. Riot has tightened coach communication rights before; Valve remains more permissive toward third-party data.

If you run an organization, ask yourself one question: does your advantage come from understanding the game, or from owning a door others have no key to?

And if you are a fan, remember this: every match is a confession of probability. When a tool stands behind the door of the strategy room, that confession no longer belongs entirely to the players.