Possible Techniques for Cheating in High-level OTB Chess tournaments : chess

0
17
Possible Techniques for Cheating in High-level OTB Chess tournaments : chess


Because a lot of the discourse is around “Did Hans Niemann cheat” and the current dialog about “How could one possibly cheat” seems limited to a meme about intimate jewelry, I thought I’d weigh in a bit about methods which could be used to get information from outside a tournament hall into the hands of a player at the board. This isn’t an accusation of any kind, I’m just interested from a security perspective what sort of methods could be used to transmit moves to players in chess tournaments. I think the conversation about how cheating is possible is broader and more important than trying to figure out if a given player has cheated in this instance.

I’d like to take a moment to say that I’ve never played chess at a high level, and my knowledge of the security posture of these tournaments is limited to recent coverage. These are the countermeasures that I’m aware of:

Given that I’ve only seen players being swept, and RF detection only works for active devices, it’s also entirely possible that a radio device could be brought in by a third party and passed off at some point after security checks are complete. However, I think that the discussion about radio devices is a bit obvious, and threats within the hall are easier to mitigate and control. Furthermore, if you have an agent with perfect information inside the hall, covert transmission of information between the agent and the player is probably somewhat trivial. This has been done before in high level chess by Hauchard, Feller and Marzolo. I’d like to mention some methods that could be used to transmit data from outside the playing hall to an operator with no electronic device inside the hall.

Threat Model

These are the main pieces of the threat model that I am looking at:

  • Per the discourse on this issue from GMs such as Caruana, Nakamura and Naroditsky, top players only need to be given one or two moves, representing maybe 4-8 bytes of information total. In my opinion, this almost completely negates the countermeasure of statistical analysis (eg Ken Reagan’s model.)

  • Players are allowed to leave the board and walk around the playing hall

  • Players are allowed to take unsupervised bathroom breaks.

Methods:

There are 3 potential methods that I can think of to covertly obtain data from outside the playing hall.

  • Morse Code (or similar) Via light. There are many examples of methods using morse code to transmit data covertly and/or out of airgapped systems. For our purposes, a laser could be used to transmit morse code from far outside the playing hall via a window, skylight, or other opening in the building. This laser could be a tiny dot on a wall that would be almost unnoticeable to someone not looking for it. In order to render the laser completely invisible to other players, perhaps a simple glass or plastic lens could be used to reveal an out-of-visible-spectrum point on the wall, or it’s possible that a lens could be worn to protect the eye as a laser was shined directly into the eye. Low spectrum lasers could theoretically be shined directly and repeatedly into the eye with no eye damage. It’s possible there’s low-spectrum laser technology that wouldn’t cause flash blindness and just give a direct visual indicator only in the field of vision of the recipient, but I’m not appraised of this technology.

  • Morse code (or similar) via sound. Subsonic sound can be used to transmit data over distance. One example that uses household objects is shown here, though a more sensitive system may be required to decode without high-speed cameras/devices. One possible example would be to have a hair of defined length placed in a bowl or glass of water (eg. in a toilet,) and vibrations at the resonant frequency of that hair would produce visible ripples around the hair. Another method involves distant noise such as car horns honking, bass from a passing vehicle blasting dubstep, construction noise, etc. Imagine a more clever version of the Charles Ingram scandal from Who Wants to be a Millionaire.

  • Semaphore/sign language/smoke signals/etc. This one is obvious on its face, but can actually work over great distances as well. Particularly if players are allowed to leave the table and look out a window, visual information could be transmitted covertly in countless ways limited only by the budget and creativity of the operatives. Cars tapping on brakes a set number of times, reflections in a window across the street, digital billboards with realtime data hidden in them, etc. etc. etc.

Conclusions:

As prize money increases, the incentive (and effort) to defeat the security measures put in place by tournament organizers will necessarily increase. I’m a computer systems expert, not a security expert. There are probably a LOT of ways that I’m not thinking about to exfiltrate data to a recipient with no device. Frankly, if high tech solutions are brought to bear on this problem, it’s not looking good for classical chess. From an anti-cheat perspective, Blitz and Rapid are much better formats both because the time control limits many data transmission techniques and because players cannot feasibly leave the table, which makes the threat model magnitudes narrower. If chess clubs are not already doing so, in my opinion a few common sense measures should be taken:

  • Players should not have access to windows, even escorted. Windows should be papered over with some type of material that doesn’t allow any non-diffuse light to enter.

  • Ideally, the playing hall should be a Faraday cage. Otherwise, cell jammers and RF jammers should be used in the hall.

  • Nobody other than players and tournament staff should be allowed anywhere near the playing area.

  • Sightlines of players to other humans or potential data sources should be considered by tournament organizers and arbiters at all times.

  • A 30 minute delay should be used for all broadcasts of classical time controls.

  • High fidelity audio and video recordings should be made of not just the games, but also the players and the entire tournament play area.



Source link