AI Solly outbluffs elite humans at Liar’s Poker

Kari Jaaskelainen

06 Nov 2025 — 1 min read

AI that thrives on bluffing

Researchers built Solly, an AI that learns Liar’s Poker—a fast, bluff-heavy game with many players and hidden info.

Instead of studying only two-player poker, Solly tackles truly multi-player dynamics. Trained via self-play with a model-free, actor-critic deep reinforcement learning approach, it taught itself when to bid, call, or bluff.

Performance: Won over 50% of hands and earned positive equity in both heads-up and multi-player matches.
Against people: Played at elite human level and wasn’t easily exploitable by world-class players.
Against AI: Outperformed large language models, even those designed for reasoning.
Strategy: Discovered novel bidding patterns and used effective randomization to stay unpredictable.

Why it matters: Liar’s Poker pushes AI beyond perfect-information games into messy, human-like uncertainty and deception—a step toward systems that can reason, negotiate, and make decisions with incomplete data.

Paper: arXiv:2511.03724

Paper: http://arxiv.org/abs/2511.03724v1

Register: https://www.AiFeta.com

#AI #ReinforcementLearning #SelfPlay #GameTheory #DeepRL #MultiAgent #LiarsPoker #Poker

Automating GDPR Compliance: A Roadmap for Companies and Law Firms

GDPR compliance is more than checkboxes. A new roadmap from the Privatech project shows how automation and machine learning can help companies and law firms assess—and even generate—privacy compliance. * Shift the focus to data processors’ real workflows: drafting policies, mapping data uses, documenting decisions. * Break compliance into machine-ready

FPGAs for Faster, Leaner Deep Learning: A Review of CNN Accelerators

Deep learning drives image search, robots, and medical scans. Most systems lean on CPUs and GPUs. This review asks: what if we run convolutional neural networks (CNNs) on FPGAs—reconfigurable chips you can tailor to the model? * Why FPGAs: custom dataflows, low latency, and strong energy efficiency—great for cameras,

Dynamic-K: Recommendations That Know When to Stop

Most apps show a fixed number of “top” items—say 10 movies or 20 products—assuming there are always enough good options. But that’s not always true: sometimes there are few relevant items, or some users are extra picky. The result? Filler recommendations. Dynamic-K flips the script. Instead of

Teaching chatbots to stop contradicting themselves (DECODE)

Teaching chatbots to stop contradicting themselves Ever had a bot say one thing, then the opposite a few turns later? This study introduces DECODE—a new task and dataset for spotting contradictions in everyday conversations, drawn from both human-human and human-bot chats. * New data beats existing natural language inference (NLI)

AI that thrives on bluffing

Read more

Automating GDPR Compliance: A Roadmap for Companies and Law Firms

FPGAs for Faster, Leaner Deep Learning: A Review of CNN Accelerators

Dynamic-K: Recommendations That Know When to Stop

Teaching chatbots to stop contradicting themselves (DECODE)