Al and fair play in online gaming

When Parliament enacted the Promotion and Regulation of Online Gaming Act, 2025 (PROGA), it resolved years of debate by prohibiting all online games involving money, whether of skill or chance. The Act answered one question, but it left another hanging in the air. If money games are banned, what becomes of the vast universe of games without money – the e -sports tournaments, casual mobile titles, and social platforms that occupy the time of nearly 488 million Indians? How safe are these spaces when Artificial Intelligence, often invisible to the player, now shapes almost every aspect of the experience?

The anxiety is not abstract. AI systems are now embedded in matchmaking, progression design, and even in the form of “black-box agents” that play against users. These opaque, non-interpretable models are agents whose internal logic cannot be explained easily to the end user. Players are left to wonder whether they are competing against other humans or against an algorithm fine – tune to manipulate engagement. Courts abroad have recognised this as a legal problem. In MDY Industries , LLC v. Blizzard Entertainment, Inc., 629 F.3d 928 (9th Cir. 2010), the U.S. Ninth Circuit upheld liability against a company selling bot software for World of Warcraft. In Blizzard Entertainment, Inc. v. Bossland GmbH [2017] EWHC 252 (Ch), the English High Court granted injunctions and damages against cheat-software developers, while the German court imposed multi-million euro penalties on the same company. Poker networks, operating in regulated jurisdictions, have gone further by banning bot accounts and reimbursing affected players.

Advertisement

The principal emerging from these precedents is clear: undisclosed automation is not innovation, but manipulation, and it falls squarely within the reach of law. Indian statutes already provide fragments of protection. The Consumer Protection Act, 2019, read with the Central Consumer Protection Authority’s Guidelines on Prevention of Dark Patterns, 2023, prohibits deceptive design which includes algorithmic nudges and manipulative difficulty settings. The Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, as amended in 2023, brings online gaming intermediaries within a regime of due diligence and transparency, a framework that could be extended to mandate AI disclosure.

The Digital Personal Data Protection Act, 2023 constrains the collection and processing of telemetry data, requiring necessity and proportionality in model training and anti-cheat systems. Most importantly, PROGA itself creates an institutional foundation. Section 4 of the Act provides for the establishment of a Central Online Gaming Authority, empowered to regulate, license, and frame rules for the sector. While its immediate mandate is to oversee compliance with the prohibition on money games, the Authority has the power to set standards for safety, integrity, and fair play in non-monetary games as well. Comparative law provides persuasive models. The EU Artificial Intelligence Act (2024) requires transparency whenever users interact with AI and prohibits manipulative AI systems rules that would cover hidden bots in competitive play.

The EU Digital Services Act (2022) bans certain dark patterns and mandates disclosure of a logarithmic recommender system. The UK Online Safety Act (2023) imposes statutory safety duties on platforms with interactive features, including games, and requires child-specific safeguards. South Korea’s Game Industry Promotion Act directly empowers regulators to address speculative mechanics and automation . European debates on loot boxes, from Belgium’s gambling classification to the Netherlands’ narrower interpretation, demonstrate the regulatory seriousness with which game design choices are now assessed. India need not copy these models verbatim, but it can adapt their principal.

The new Authority could require that platforms disclose when a player faces an AI-driven opponent, establishing transparency as the baseline. Certification mechanisms akin to RNG audits in gaming jurisdictions such as GLI-19 or eCOGRA could evolve into an “anti-bot and fair-match” seal, audited annually. Dark-pattern prohibitions should be extended into gaming contexts, covering both interface design and algorithmic progression loops. Child-first defaults, such as disabled chat or periodic time-out prompts, can draw from the UK model. Due-process standards, already familiar under the IT Rules, should apply to player suspensions and bans, ensuring notice and appeal.

Finally, a shared “integrity hub,” overseen by the Authority, could enable platforms to exchange data on both signatures and enforcement actions. Framing the problem in legal and technical terms also ensures the conversation is not dismissed as mere sentiment. The issues are thus algorithmic accountability, model interpretability, and liability frameworks. In AI governance language, the demand is for “explainability,” “auditability,” and “ bias detection” in game design. These principles, when embedded in law, prevent games from degenerating into black-box contests where outcomes are predetermined by hidden code. The PROGA Act has settled the money question. But money was never the only axis of harm. With its new statutory Authority, India has the opportunity to craft the first comprehensive user-safety code for digital users. If it succeeds, players will compete on merit, not against stealth code, and India will set the global benchmark for algorithmic fair play in digital entertainment.

(The writer is a Delhi-based lawyer who heads a think-tank, EWA Centre.)