Jack Williams, iTero and GIANTX: AI Coaching Is Touching the Fairness Boundary of Esports

Thursday night in Busan, after my market-analysis broadcast, I stayed behind to finish reading the Jack Williams interview about iTero. This should have been the kind of content I scroll past: a small B2B piece with no transfer fee, no release clause, no domino to trace. But two section headings made me stop. The first: working exclusively with GIANTX and the likelihood of being copied. The second: AI-assisted cheating. In Vietnam, we are used to reading esports news as results: who won, who reached Worlds, who was eliminated in groups. But the tooling layer behind the stage is what shapes results over the next two to three seasons. The loudest noise is often where the most important signal hides — and this time the signal was not in Jack Williams's answers, but in the choice to place those two headings side by side in the same piece. There would be nothing to discuss if iTero were just statistics software. But when a tool is framed by an exclusivity contract rather than by its features, the story leaves technology and enters governance. And in governance, nobody reads the contract carefully until someone is left behind. For readers unfamiliar with European esports, let me rebuild the context. iTero is an AI-driven analytics and coaching platform for professional esports teams. Jack Williams is the person behind it. The interview revolves around three axes: what the product does, how far the GIANTX partnership goes, and how close AI is edging to the cheating line. GIANTX is a European esports organisation operating mainly in the LEC — the EMEA League of Legends league run by Riot Games. Analysts note GIANTX formed from a merger between Excel Esports and Giants Gaming, but that detail needs verification, as a different entity under the same name may exist. I mention it to mark my confidence level, not to assert it as fact. On timing, the interview anchors itself to an event: the interviewer mentions Natus Vincere lifting the Aegis of Champions at Gamescom "14 years ago". Natus Vincere won The International for the first time in 2026, also at Gamescom. The subtraction yields 2026. That is arithmetic inference from the article's own wording, not disclosed information. At the tactical layer, one difference I always teach my students: Valve and Riot run their update cadences in opposite directions. Valve ships large, infrequent patches that overturn systems; between them sit long stretches of stability. Riot patches every two weeks, giving every machine-learning model a short shelf life. Both are called "AI coaching", but their value inverts by title: in slow-patch titles the value lies in depth of historical modelling; in fast-patch titles the value lies in speed of detecting meta shifts. A single product advertised identically across both titles is a suspicious sign. And here is the point the interview touches but does not dig into: in a closed league like the LEC, there is no relegation. Every member is a permanent member. A structural advantage held by one team is not competed away over a season — it accumulates across seasons. That is why exclusivity carries different weight in a closed league than in an open system. For comparison, I always use the LCK as a reference familiar to Vietnamese readers. Korean esports has a tradition of tight governance, with monitoring systems and competitive-conduct rules written into formal documents. But even there, the question of which analytics tools are permitted, and who may use them, has no unified answer. What is not written into law is exactly where exclusivity deals reside. There is one detail I noticed when re-reading the source, and it caught my attention more than the content itself. Most of the text is not about Jack Williams, and not about iTero. It is about the person asking the questions. I do not treat this as an editorial error; I treat it as evidence of how this industry operates. When a B2B product lacks enough data to prove itself, people fill the gap with the storyteller's personal story. That is not a moral judgement. It is a structural observation. And that structure tells me the real maturity level of the esports tooling market. Now to the core, which I analyse on structure rather than product data — because the interview discloses no figures. First, the commercial logic. In a closed league, an exclusivity deal is not a reward for better technology; it is how you turn easily copied technology into an asset that cannot be copied. If the algorithm can be rebuilt by a rival in six months, the only sellable advantage is access. Rumours are the surface. The system sits underneath. Here, the surface is the AI story; the system is the exclusive distribution contract. Look at the first heading itself: "the likelihood of being copied". The fear of being copied shows they know their product can be reproduced. When an asset is easily reproduced, the only way to preserve value is to lock distribution. That pushes the strategy from research and development into exclusive partnerships. And when the strategy shifts that way, the product no longer competes on quality but on a locked customer network. In the transfer market I have seen this pattern many times. A club does not keep a player by paying the highest wage, but by inserting a release clause expensive enough that nobody touches it. A software tool works the same way: if it cannot retain users by quality, it retains them by contract. Same logic, different currency. Second, the exclusivity layer. An exclusivity deal has two signatories but many parties who bear the cost. That cost does not sit on GIANTX's balance sheet. It sits with the nine other teams in the league, with young scouting teams that cannot afford the tool, and with regions like Vietnam where a tier-two team must choose between rent and software licences. A failed contract is an open diary — but here what is left open is someone else's exclusivity deal, and the teams outside the game will never read that diary. I have followed regional leagues long enough to see this repeat. When a team holds a data advantage, that advantage does not show in a single match. It shows in game three of a BO5, when the opponent has run out of options and the other team still has an unused layer of analysis. That is an advantage you cannot see in the regular-season standings, but you can see it in deciding-game win rates. And deciding-game win rates are exactly what this interview does not disclose. Third, the cheating layer — the second heading. This is where I want to separate things clearly. Real-time in-game assistance is already clearly banned in every major title. So the real debate is not there. The interesting grey zone lies between games in a BO3 or BO5 — where coaches are allowed to enter, adjust, and speak with players. If an AI tool summarises the previous game's data and suggests tactical adjustments during that break, it is doing exactly what a coach does, only faster and without fatigue. The line between "analytics assistant" and "an undeclared sixth player" blurs precisely at the between-game break. I have watched enough matches to know this is not hypothetical. In leagues where equipment and software rules inside the playing area are loose, the between-game break is when information flows most freely. In strict leagues, that break is capped by how many people may enter and how much analysis time is allowed. A tool running on a coach's personal laptop breaks fewer rules than one running on a match machine. That is a structural loophole, not a technical fault. Fourth, the verification layer. The interview discloses no sample size, no evaluation methodology, no prediction accuracy. That turns every performance claim into a black box. In the transfer business I hold a principle: a number without a source is not a number, it is an advertisement. Here too. A tool that says it improves preparation but does not say by how much, across how many matches, against what control group — that is marketing, not evidence. I do not deny the value of analytics tools. I deny the way they are sold on faith rather than data. And in a market where the buyers are teams under weekly performance pressure, faith is an easier product to sell than statistics. This is the section I want to give to the counter-intuitive angle, because I believe this is where the interview exposes its own blind spot. The two chosen headings — exclusivity and cheating — frame the issue two ways: a commercial frame and an integrity frame. But a third frame sits between them, and it does not appear: the league-fairness frame. Public debate pours into "is AI cheating" and "will my algorithm be copied". But the question nobody asks is: "is one team allowed exclusive ownership of a tool that nine others lack, in a league with no relegation?" When a closed league permits tool exclusivity, it is implicitly choosing to permit preparation inequality. No decision is announced, no rule is broken — the inequality forms itself as a consequence of silence. This is the classic blind spot: people defend the rules of play on the field while the rules behind the field go unwritten. And here I must be careful with my own frame of reference. I live and work in South Korea, where the esports governance tradition is far stronger than in most regions. If I used that standard as my yardstick, I would easily conclude that every Western exclusivity deal is a problem. But the right question is not "which region governs tighter", it is "what is this ecosystem optimising for". An ecosystem optimising for commercial revenue sees exclusivity as an innovation incentive. An ecosystem optimising for open competition sees exclusivity as a barrier. The same deal, two conclusions, depending on the system's goal. There is one more detail about patch cadence that I consider more important than all the rest. If iTero sells to both slow-patch and fast-patch titles with a single advertising line, then either the product is genuinely neutral to a suspicious degree, or the pitch is hiding that it is only strong in one. In a two-week patch environment, value lies in speed. In a slow-patch environment, value lies in depth. No product optimises for both at once without a trade-off. And the party paying for that trade-off is usually the team buying the tool, not the seller. This is why I call this story an integrity test, not a technology pitch. The first to know is not necessarily right, but is the one who creates the shock. And the shock here is not that AI can coach. The shock is that a contract knows how to lock the road. I do not think iTero did anything wrong. The esports tooling market is at a stage where every player does what is rational for themselves. But precisely because of that, the next domino will not come from the teams. It will come from the organisers. When a tool can affect competitive outcomes, league operators will soon have to choose one of two paths: mandate equal access for all teams, or restrict the tool. This is the road that in-game coach communication regulation already travelled — from permission, to limitation, to outright bans or standardisation. The same logic will apply to the analytics-tool layer, just a few seasons slower. For Vietnamese readers watching from afar, this is the moment to take notes. Not because one European exclusivity deal will change domestic leagues overnight, but because it draws the map of what is coming. When the tooling layer becomes a layer that can be locked by contract, teams without contracts will have no tools. And in the transfer market there are no accidents, only things we have not read carefully enough. I still sit down after every broadcast to read the interviews others scroll past. I started taking notes because one deal fell apart, and I have kept taking notes ever since. Not because I believe everything I read, but because I want to know who is signing with whom, and who is being left out. The question I leave tonight is simple: if the preparation tool becomes exclusive property, what is left in the league to distinguish tactics from budget?