Join The Community
Square Enix has launched its AI-driven reboot of arguably its most influential game, but it's an unbelievable failure as both a classic revival and an AI experiment.
Last week, the publisher announced the impending release of Square Enix AI Tech Preview: The Portopia Serial Murder Case. The original Portopia, a text-driven adventure game about solving a murder, is one of the most influential titles in the Japanese gaming canon, yet despite that it had never officially been released in English. The idea of getting an official translation alongside an innovative use for an AI language model to replace the old text parser seemed potentially exciting.
That excitement fell to pieces pretty much the instant I started typing into the new Portopia's text box. As with the original game, you don't directly control your character, instead giving instructions to your assistant, who carries them out on your behalf. So you might say "ask around" to have him question local passers-by, or say "tell me about the victim" to get some details on the character.
The issue is that the AI is barely capable of responding to all but the most narrow of questions and commands. If you've ever played an old-school text adventure and had to fight to figure out what vocabulary the computer would actually understand, you're starting to understand the frustration here. It's just about a thousand times worse in the new Portopia, because instead of needing to learn a handful of specific commands the game recognizes, you've got to work around the vagaries of the algorithm trying to understand your words.
If you type something the AI doesn't recognize - which is going to be about 95% of what you input - you're going to get one of a tiny handful of canned responses from the game, like "I'm not sure what to say about that," "hm," or "maybe we should focus on the task at hand." Not exactly the AI-driven future we've been told about.
The strangest thing about this new version of Portopia is that there's no AI-generated dialog in response to your input. AI writing is controversial - with good reason - but it seems like if you're going to put out an "AI tech preview" you'd want to go all the way with it. Having an AI companion to give you hints as you stumble toward the next story trigger could even be the detail that makes the whole game click into place.
"This tech preview originally included a function based on Natural Language Generation technology, where the system would generate natural replies to questions that did not have a pre-written response," Square Enix says on the game's Steam page. "However, the NLG function is omitted in this release because there remains a risk of the AI generating unethical replies. We thank you for your understanding. We will consider reintroducing this function as soon as our research succeeds in creating an environment in which players can enjoy the experience with peace of mind."
Yes, the fear of "unethical replies" is what's keeping any of this from making sense. Most AI chat tools implement some sort of filter to stop the responses from getting too horny or offensive - ChatGPT was very insistent on defending the honor of Pokemon icon, Ash's mom - but I continue to question why this game exists in the first place if it's going to exist in such a compromised state. It's a free preview intended to showcase the future of AI tech in gaming, but if this is what we can expect then you can count me out.
