The video game industry’s relationship with its audience is being rewired. According to a recent survey by games research firm Fancensus, 55% of its gamer panel have already used artificial intelligence to find information about games. This rapid adoption positions AI as a critical intermediary, shaping how players discover new titles, interpret major events, and form their perceptions. An in-depth case study of The Game Awards 2025 illustrates this evolution, revealing how AI platforms like ChatGPT, Gemini, and Perplexity have moved beyond being simple curiosities to become powerful gateways that filter, amplify, and frame the news and narrative around gaming’s biggest announcements. This analysis exposes a new battleground for visibility, compelling publishers and developers to develop AI-aware strategies alongside traditional marketing efforts.
AI as a Media Channel: A New Methodology for Discovery
To quantify this emerging influence, Fancensus treated AI platforms as measurable media channels. Their study, AI Engine Observations – The Game Awards ’25, involved querying three leading AI platforms with a consistent set of over 2,000 structured questions across three distinct periods: before, during, and after the live broadcast. This methodology tracked pre-show rumours, captured real-time reactions to announcements and winners, and analyzed post-show narrative consolidation. By cycling questions every five minutes during the event and continuing regular queries afterward, the study provided a granular view of how AI interprets and prioritizes information in near-real time, effectively creating a secondary, AI-driven layer of content on top of the official event.
Pre-Show Analysis: The AI Rumor Mill in Action
In the days leading up to The Game Awards 2025, the AI platforms surfaced a broad spread of rumoured titles. High-profile franchises like Resident Evil Requiem were consistently highlighted, reflecting the intensity of online discussion. However, the accuracy of these pre-show predictions remained modest. A significant finding was the AI’s frequent blurring of lines between games that were officially announced, merely nominated, or simply wishlisted by the community. This tendency underscores a key risk: speculative conversation can easily be processed and presented by an AI interface as perceived fact. For publishers, this period represents both opportunity—legitimate teasers can gain rapid traction—and peril, as inaccurate expectations can become cemented in the AI-mediated chatter.
Top Rumoured Games According to AI
When repeatedly asked, “What games are rumoured to be announced at the Game Awards 2025?” in the four days preceding the show, the combined responses from all three AI platforms most frequently mentioned titles like Resident Evil Requiem, Armored Core VI: Rubicon Awakening, and Metroid Prime 4. This list, derived from aggregated data, reveals which rumours had the greatest velocity and penetration within the AI’s source ecosystem, offering a snapshot of pre-show hype as seen through an algorithmic lens.
Live Show Dynamics: Shaping the AI-Driven Winners’ Circle
Once the live broadcast began, the AI platforms quickly converged on a handful of defining titles. Clair Obscur: Expedition 33, which won multiple awards, unsurprisingly became the dominant reference point across all queries. Questions about the “most exciting” reveal or “biggest surprise” repeatedly elevated the same cluster of games. The timing of a title’s appearance in the show also proved crucial; games featured earlier had more time to be captured and resurfaced in subsequent AI responses, strengthening their perceived prominence. This real-time processing meant the AIs effectively created their own hierarchy of importance, a secondary lineup layered atop the official running order, influencing what immediate post-show discussions might focus on.
Post-Show Consolidation: How AI Frames the Legacy
In the 48 hours following the event, AI responses shifted from real-time capture to narrative consolidation. The platforms began positioning certain games as emblematic of the show’s overall impact and legacy. Surprise reveals and polarising moments, such as the finale for Highguard, were amplified in AI summaries. Conversely, some official award winners received less AI attention than their trophy haul would suggest. This indicates that memorability in AI outputs is driven as much by discourse intensity, distinctiveness, and cultural buzz as by formal recognition. Mentions of personalities like hosts and standout performers followed a similar pattern, becoming shorthand for the event’s cultural impact in AI-generated recaps.
Divergent Editorial Behaviours Across Platforms
Despite drawing on overlapping sources, the three AI platforms exhibited distinct editorial instincts. One platform relied heavily on a small set of established reference sites like Wikipedia. Another distributed citations across a far broader mix of specialist gaming outlets, general news, tech media, and user-generated platforms such as Reddit and YouTube. Notably, official publisher websites were referenced only in a fragmented manner. These sourcing differences led to subtly different hierarchies of games and publishers listed across the platforms. Consequently, a title’s perceived prominence can vary significantly depending on which AI a player queries. This highlights the importance for publishers to ensure accuracy and completeness of details on their owned websites, as these can serve as key sources for AI.
The Source Gap: A Quantitative View
The study quantified this divergence. During the live show, Gemini cited a significantly higher number of unique sources compared to ChatGPT, referencing nearly 100 more distinct outlets across the repeated question cycles. This demonstrates that not all AI platforms curate information equally, and understanding which sources are prioritized by each can be key to ensuring messaging cuts through.
Identifying AI Strengths and Critical Blind Spots
At an aggregate level, the AI platforms proved competent at identifying core winners, major announcements, and broad community sentiment. However, the study also revealed clear blind spots. Pre-show rumours were often misclassified, and some surprise reveals were under-weighted in AI summaries. More concerningly, instances of hallucination occurred, with at least one platform reproducing outdated or completely incorrect information about the event’s format, confusing details from earlier years. For stakeholders, this underscores a critical point: AI-derived summaries can be directionally useful for gauging perception, but they are not a substitute for verified, official reporting and carry inherent risks of propagating inaccuracies.
Strategic Implications for the Gaming Industry
The findings have profound implications for both game publishers and platform holders. For publishers, AI optimisation is now a new frontier. Being “top of mind” for AI platforms depends on a combination of clear metadata, sustained coverage across a diverse range of reputable outlets, and strong post-show discussion, rather than on awards alone. Understanding and monitoring the AI-amplified narrative is crucial, as aggregate perception from across many sources can result in negative or inaccurate narratives dominating responses and becoming “baked in.” Early measurement of sentiment and source influence is key to maintaining control of the narrative.
For platform holders like Sony and Microsoft, the study reveals that questions such as “What was announced for PS5?” are now being answered first by AI systems. The data showed that the combined AI responses mentioned 43% more PS5 titles and 68% more Xbox Series titles than were actually featured in the show. This indicates a high level of inferred or incorrect platform attribution, making the accuracy and clarity of platform-exclusive messaging within AI channels a commercially significant concern.
Building an AI-Aware Launch Playbook
Ultimately, Fancensus’s analysis positions AI searches and chatbots as a new class of soft curator: powerful, non-official filters that significantly influence which games players hear about and how they remember industry events. This necessitates a shift in strategy. As more players default to conversational AI interfaces instead of traditional search engines, treating these platforms as measurable, optimisable touchpoints will become integral to launch planning. A comprehensive “AI visibility” programme, tracking a title’s presence and portrayal in AI responses throughout a campaign’s lifecycle—from rumour to long-tail discovery—is fast evolving from a forward-thinking experiment into a core component of modern game marketing and audience engagement.
The transformation is already underway. The conversation between publisher and player is no longer direct; it is mediated, interpreted, and reframed by intelligent systems that learn from the entire digital discourse. Mastering this new layer of communication—understanding its rhythms, its biases, and its points of leverage—is no longer optional for those seeking to capture attention in an increasingly crowded and algorithmically guided marketplace.
