AI Insight
Researchers developed PUBG Ally, an AI agent that plays the video game PUBG: BATTLEGROUNDS alongside human players while communicating through voice. The system combines language model-based reasoning for decision-making and natural conversation with a fast control layer for real-time gameplay actions. The team collected data from nearly 39,000 gameplay sessions with real players and used iterative training and player feedback to refine the agent's performance, with survey results from 141 countries showing net positive recommendations of 25.1 percentage points.
Why it matters
This work demonstrates a practical integration of conversational AI with real-time interactive systems, addressing challenges like low-latency execution, safety guardrails, and synchronization between speech and actions. The approach of using actual player interaction data and preferences to iteratively improve AI teammates could inform development of AI assistants in gaming, collaborative work environments, and other human-AI teaming applications.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Abstract: We introduce PUBG Ally, an embodied agent for PUBG: BATTLEGROUNDS that can reason, act autonomously, and play alongside players as a voice-enabled teammate. Building such a teammate requires combining two difficult capabilities: it must perceive and respond to a constantly changing game world under strict latency constraints while interacting naturally with players, keeping its speech synchronized with its actions. Ally therefore combines agentic tool use with real-time game control. A language-model agent uses a controlled interface to inspect game information, interpret player speech, maintain context, decide what to say, and issue high-level action choices that steer a faster control layer for movement, combat, and recovery. Because the player’s and Ally’s speech and actions continually shape each other and the course of the match, training requires data from actual gameplay. We therefore collect data across nearly 39k sessions in which real players play alongside Ally, recording gameplay, player speech, agent decisions, tool use, actions, and player feedback, and use these records for iterative training. To evaluate teammate quality, we use player feedback and preference comparisons to identify gaps between offline evaluations and player preferences, and iteratively refine the evaluation criteria. Deploying Ally in live service further requires low-latency on-device execution and safeguards for player-facing communication, which we address through model compression, context compaction, targeted safety training, runtime guardrails, and memory redaction. During the live service, we surveyed players in 141 countries. Among respondents whose play with Ally was confirmed in game records, positive responses exceeded negative responses by 25.1 percentage points when asked whether they would recommend Ally, with players describing Ally not only as a tool but also as a teammate or companion.
Source: PUBG Ally: A Conversational Embodied Agent as an AI Teammate