Swery65’s new horror-action game, Hotel Barcelona, has just added a generative AI label after failing to disclose the fact at launch.
Hotel Barcelona, a collaborative effort from Swery65 and Suda51, is in the spotlight for the awkward handling of disclosure regarding its use of generative AI. A few days after its September 26, 2025, release on Steam, the game’s store page has been updated with an AI generated content disclosure that was not present at launch, raising concerns among fans about transparency and compliance with Steam’s own rules.
According to the updated Steam page, two elements in Hotel Barcelona originally made use of AI. First, VHS tapes found in the late stages of the game contained voiceovers produced with a text-to-speech service. Second, a logo featured on those same tapes had been completed with generative fill AI. The developer has further mentioned that these features will be removed in an upcoming patch, and will be replaced with non-AI content.
The issue goes beyond just the assets themselves, however. Fans on the Reset Era forums were quick to highlight that the absence of disclosure at launch potentially violated Steam’s Terms of Service, which require developers to clearly flag the use of AI-generated material. One player remarked, “I’d hate if I had bought the game and found out about the AI usage after playing more than two hours, thus unable to get a refund.” Others expressed disappointment on a creative level, arguing that using generative tools undermines originality, which gave creators like Swery and Suda prominence in the first place.
Meanwhile, some questioned Valve’s handling of the situation, asking why a title could go live without an AI disclosure in the first place if rules were already in place. One player suggested that Steam should allow extended refunds for customers who purchased the game before the label was added, saying, “They should honor refunds for some period, 48-72 hours at least, for people who purchased.”
While the developer’s decision to remove AI-generated content is a step in the right direction, Valve’s handling of the situation remains a question mark for future violations pertaining to disclosure over the use of generative AI in games.
