FRITZ 20, a False FIDE Presidency Headline, and the Limits of Trust in Chess Media

A question from a fourteen-year-old

On the afternoon of 29 August 2026, the ceiling fan in a chess club on Nguyen Chi Thanh Street in Da Nang turned slowly enough that I could hear a knight being placed on the board. Phuc, fourteen years old, pushed his phone across the table: "Uncle, is Timur Turlov the FIDE President now?"

On the screen was a bold headline: the FIDE President meets the President of Cyprus in Nicosia. A few lines down, the article drifted into a promotion for FRITZ 20, the chess software, with three slogans lined up side by side: your personal chess trainer, your toughest opponent, your strongest ally.

I opened my own phone, typed FIDE, went to the official site. The portrait of the sitting president was still Arkady Dvorkovich. No press release announced a handover. No resolution, no extraordinary congress, no line placing Turlov in a leadership role.

It took Phuc ninety seconds to believe something false. It took me ninety seconds to check. The distance between two people sitting at the same chessboard is exactly one search query.

In 2026, building the video series that broke down Nguyen Thi Oanh's 1,500m race at the SEA Games, I learned something about young readers: they are not afraid of data, only of fake data. Young audiences read fast, and false news runs a couple of laps ahead of true news.

Who actually holds the FIDE presidency

Arkady Dvorkovich was elected FIDE President at the FIDE Congress in Batumi in October 2026, defeating Georgios Makropoulos 103 votes to 78. Four years later, at the Chennai Congress in August 2026, he was re-elected with 157 votes against 16 for Andrii Baryshpolets. Those are traceable milestones, with minutes and delegate lists.

Timur Turlov is a Kazakhstan-born businessman associated with Freedom Holding Corp, a Nasdaq-listed group. His name appears in chess through sponsorship: events and federations in Kazakhstan have received money from funds linked to Freedom Holding, and private capital has pulled international chess events toward Almaty. All of that is real, and it deserves a proper article.

Sponsorship is not a presidency. A patron does not automatically become the head of the world federation. If Turlov ever does run, that event will come with a press release, a nomination deadline and a list of candidates, not a headline that appears first while the body text underneath sells a piece of software.

The structure is what caught my eye. The headline is about power. The body is about a product. The two are unrelated, joined by a colon. In my trade we call that an authority hook: borrow a big name to open a door, then push through the door whatever needs selling.

The Nicosia meeting, if it happened, is not unusual. The sitting President of Cyprus is Nikos Christodoulides. National chess federations routinely invite heads of state as guests of honour for an open tournament or a school chess programme. But such a meeting belongs in a short sports brief, not on a front page, and certainly not as a pretext to sell software.

FRITZ 20: three slogans and not one specification

The Fritz line is no newcomer. Frans Morsch wrote the first version in 2026 within the ChessBase ecosystem. In 2026, X3D Fritz drew 2–2 with Garry Kasparov in New York. In November and December 2026, Deep Fritz beat Vladimir Kramnik 4–2 in Bonn. Those games carried the chess machine from the laboratory to the front page, and I still keep my handwritten notes on the sixth Bonn game, where Kramnik let a diagonal slip.

So when a new version appears, the right question is not "is it a revolution" but "how strong is it, how is that strength built, and who measured it".

The three slogans answer none of those. Personal trainer: training what, following which curriculum, with spaced repetition or not. Toughest opponent: tougher than Stockfish by how much, at what time per move. Strongest ally: built on which architecture, running on which hardware.

A serious engine release, covered seriously, carries six categories of data. First, architecture: an NNUE-style neural network or classical search, hybrid or not. Second, an Elo estimate with a named comparison target and specific hardware conditions. Third, results in automated arenas such as CCRL or TCEC, where everyone can inspect the same data. Fourth, self-play game counts and time controls. Fifth, opening book, endgame tablebase and database size. Sixth, and for me most important, the actual training functions: game analysis, blunder detection, error classification, exercises generated from the user's own games, progress tracking month by month.

All six were absent.

I have watched enough TCEC to know one thing: in the engine market, the word "revolution" is currency anyone can print. Stockfish is free, open source, and almost always at the top of the lists. Leela Chess Zero is strong in a human-like style and needs a GPU. Komodo Dragon sells a commercial build with its own neural network. Fritz does not compete at the summit of raw strength. It competes on interface, on training workflow, on the fact that a beginner can open it and understand what to do next.

That is a real market and a real advantage. And it does not need a false headline about the FIDE presidency to sell a copy.

The line about serving both absolute beginners and tournament-level players is a familiar commercial hedge: maximum coverage. It is not a lie, but it says nothing. Buyers are given nothing with which to judge for themselves, and that is the actual problem.

The funnel: authority hook, marketing payload, a very familiar error pattern

Articles like this do not appear by accident. They follow a mould: a powerful name in the headline, a vague event in the lede, a product in the final paragraph, and no named source anywhere. No author. No publication date. No original press release. Nobody accountable for the false claim.

There is another possibility I am obliged to raise: the piece was machine-generated. The error pattern of calling a businessman the FIDE President is typical of a language model, which knows Turlov is connected to chess, knows FIDE has a president, and merges the two into a sentence that reads smoothly. So smoothly that nobody checks.

The verification routine I use takes ninety seconds, and I asked Phuc to adopt it for every chess article from now on. One: go to the official site of the organisation named, find the leadership page. Two: find the original release and compare dates. Three: find at least one independent second source, not a copy of the article being read. Four: if there is no author and no date, drop the credibility level close to zero.

Those four steps cost less than a tournament ticket. They are also chess skills: before trusting a move, check the variation.

The evaluation bar has become the new divination

This is where I want to be blunt, because it is the blind spot of an entire generation of young players.

It is the same disease as the false headline, only subtler. When software shows an evaluation bar at +2.3, a fourteen-year-old feels he understands the position. When software reports an average centipawn loss of 4.7, he feels he understands the game. But that index measures distance from the machine's line, not human understanding. It says the last move departed from the optimal choice. It does not say why, does not give the plan, does not deliver what a coach must deliver.

I have seen the same thing in another sport. Heat maps in football and basketball have become a storytelling substitute for watching the game. People look at a red cloud and pronounce on a player's role inside a tactical system, when the red cloud only says where the player stood. Data becomes the new divination: the more colours, the fewer questions.

In chess, the evaluation bar is the horoscope of the age. It delivers certainty without granting access to reasons. And a generation raised on hollow certainty will believe a false headline easily, because a false headline delivers exactly the same feeling.

There is another layer few want to see. The chess world is hungry for a saviour. Federations sell the story of "chess for the community" to sponsors, packaging women's events as a line item of corporate social responsibility, and when private money appears, the public instantly assigns a seat of power to whoever owns it. The false headline did not create that hunger. It merely sold food to it.

I count myself the stadium polymath type: writing across sports, hunting the common thread between the running track and the chessboard. This time the thread is verifiability. A track athlete does not trust someone else's stopwatch without an official. A chess player should not trust a headline without a source.

People talk about speed; I look for time. Time to run one search, read one press release, ask who did the measuring.

What I told Phuc

I did not tell Phuc the software was bad. I have no data to say that, and staying quiet when data is missing is a respectable professional choice.

I told him he had just done something many adults fail to do: he asked. He pushed the phone across the table instead of believing alone. In chess that is called checking the variation before moving. In reading the news, it is the same.

That night, when the club lights went off and one board was left open, I thought of a line I have used in my running pieces: in the silence of the stands, the future laces up its running shoes. Here, in the silence of a chess club nobody films, the future was holding a phone and waiting for someone to teach it how to check.

That skill is not in any engine update. It has to be taught, one person at a time, one afternoon at a time, by a coach willing to stay behind after practice.

And if someone in Almaty or Lausanne really is holding a seat, I will be the first to write about it, in a piece with an author, a date, sources, and the questions that person does not want to answer.