Wired has the story of a superuser cheating scandal. Professional poker is the latest community to reckon with the impact of a high-profile cyberattack. Cases where a player manages to cheat by getting unfair access to another person's cards is known as the superuser strategy, and it has popped up from time to time in the poker scene over the years. This most recent attack appears to have been a long-term campaign that may have seen millions of dollars stolen from across major online poker platforms. The story claims that the hacker exploited third-party tools allowing people to run multiple poker games at once to install a malware program called MeshAgent on specific players' computers.

Illicit software on the computers of some of the world's best online poker players allowed the accused cheater to see what cards were in the victims' hands--and trounce them at the tables. In a fitting sign of the times, the hacker's carefully laid plans were foiled by an AI agent . It was September 25, 2026, and Alex, a professional online poker player, was using his Claude AI agent to review some play statistics. After Alex's SQL database crashed and needed to be reset, the Claude agent noticed something suspicious: A program called MeshAgent was operating in the background, where it seemed to be capable of capturing Alex's screen and even controlling his entire machine. Thanks to the Claude agent's report, Alex could see the program had been installed on April 24, and not by Alex himself. Alex says the agent identified 11 time-stamped events when the program had been capturing his screen based on graphic errors it caused during screen sharing instances.

Large Language Models (LLMs) have shown remarkable progress across domains, yet their ability to perform inductive reasoning--inferring latent rules from sparse examples--remains limited. It is often assumed that chain-of-thought (CoT) prompting, as used in Large Reasoning Models (LRMs), enhances such reasoning. We investigate this assumption with creating four controlled, diagnostic game-based tasks--chess, Texas Hold'em, dice games, and blackjack--with hidden humandefined rules. We find that CoT reasoning can degrade inductive performance, with LRMs often underperforming their non-reasoning counterparts. To explain this, we present a theoretical framework that reveals how reasoning steps can amplify error through three failure modes: incorrect sub-task decomposition, incorrect sub-task solving, and incorrect final answer summarization. Based on our theoretical and empirical analysis, we introduce structured interventions that adapt CoT generation according to our identified failure types. These interventions improve inductive accuracy without retraining. Our findings suggest that effective (CoT) reasoning depends not only on taking more steps but also on ensuring those steps are well-structured.

Self-play is a technique for machine learning in multi-agent systems where a learning algorithm learns by interacting with copies of itself.

Neural Information Processing Systems http://nips.cc/

High-quality information set abstraction remains a core challenge in solving large-scale imperfect-information extensive-form games (IIEFGs)--such as no-limit Texas Hold'em--where the finite nature of spatial resources hinders solving strategies for the full game. State-of-the-art AI methods rely on pre-trained discrete clustering for abstraction, yet their hard classification irreversibly discards critical information: specifically, the quantifiable subtle differences between information sets--vital for strategy solving--thus compromising the quality of such solving. Inspired by the word embedding paradigm in natural language processing, this paper proposes the Embedding CFR algorithm, a novel approach for solving strategies in IIEFGs within an embedding space. The algorithm pre-trains and embeds the features of individual information sets into an interconnected low-dimensional continuous space, where the resulting vectors more precisely capture both the distinctions and connections between information sets. Embedding CFR introduces a strategy-solving process driven by regret accumulation and strategy updates in this embedding space, with supporting theoretical analysis verifying its ability to reduce cumulative regret. Experiments on poker show that with the same spatial overhead, Embedding CFR achieves significantly faster exploitability convergence compared to cluster-based abstraction algorithms, confirming its effectiveness. Furthermore, to our knowledge, it is the first algorithm in poker AI that pre-trains information set abstractions via low-dimensional embedding for strategy solving.

Dominance is a fundamental concept in game theory. In normal-form games dominated strategies can be identified in polynomial time. As a consequence, iterative removal of dominated strategies can be performed efficiently as a preprocessing step for reducing the size of a game before computing a Nash equilibrium. For imperfect-information games in extensive form, we could convert the game to normal form and then iteratively remove dominated strategies in the same way; however, this conversion may cause an exponential blowup in game size. In this paper we define and study the concept of dominated actions in imperfect-information games. Our main result is a polynomial-time algorithm for determining whether an action is dominated (strictly or weakly) by any mixed strategy in n-player games, which can be extended to an algorithm for iteratively removing dominated actions. This allows us to efficiently reduce the size of the game tree as a preprocessing step for Nash equilibrium computation. We explore the role of dominated actions empirically in "All In or Fold" No-Limit Texas Hold'em poker.

As a result, depth-limited search algorithms used in single-agent settings and perfect-information games do not apply.

Excessive abstraction is a critical challenge in hand abstraction-a task specific to games like Texas hold'em-when solving large-scale imperfect-information games, as it impairs AI performance. This issue arises from extreme implementations of imperfect-recall abstraction, which entirely discard historical information. This paper presents KrwEmd, the first practical algorithm designed to address this problem. We first introduce the k-recall winrate feature, which not only qualitatively distinguishes signal observation infosets by leveraging both future and, crucially, historical game information, but also quantitatively captures their similarity. We then develop the KrwEmd algorithm, which clusters signal observation infosets using earth mover's distance to measure discrepancies between their features. Experimental results demonstrate that KrwEmd significantly improves AI gameplay performance compared to existing algorithms.