cross-posted from: https://sh.itjust.works/post/70080818
I received an e-mail alert from Steam that I received the community award 💯 for one of my posts. I follow the link included, but it just takes me to the community for Big Walk. I look through my comment history, and I find it. It’s a comment I had made critical of their failure to disclose their AI usage.
So what we have learned here is that ai is bad, unless someone is a fan of the brand using it. Cool, cool, well done people.
I have never played the game and i don’t really care about it. But, from what i heard and read about it, they used ai to fix a unity bug. Before they even considered, they asked some unity engineers first, who told them it’s impossible and it can’t be done. They then asked a llm who fixed their problem in like 10 lines of code. They were upfront about it and never claimed they didn’t. Now it’s an “AI” game because of that.
If the source theg asked would’ve given a shit, non of this woud’ve happened. If the source they asked would’ve used a llm to fix tge bug and then give them the code, the game would be the same, the code would be the same, but it wouldn’t be “AI”
Assuming their claim of “well under 0.1% of Big Walk’s code [was built with AI]” is true, I personally find their reasoning acceptable:
We have some concern that voluntary disclosures like this might contribute to the normalization of this technology, or its treatment as ‘inevitable.’ We support the work of those who are trying to limit the harm generative AI is doing to videogames and our world, and apologize that Big Walk’s LLM usage may not meet their standards
This review seems counterproductive; if I were House House, I’d keep the AI usage to myself next time. I want more transparency, not less.
“well under 0.1% of Big Walk’s code [was built with AI]” sounds like someone might’ve asked Google Overview what the function for serializing JSON is. Sounds okay to me.
The detail is more extensive, yet in my opinion even less harmful. The House House devs wanted to figure out something in part of the Unity rendering engine that was causing huge slowdowns. They couldn’t figure it out, they talked to the Unity devs who told them what they wanted to do was impossible, and as a last ditch idea before giving up and removing the functionality they asked a coding agent. After a few false starts it came up with a novel solution.
I can’t fault them for that usage of a coding agent, personally.
Yeah, I probably would’ve done what they did too under such circumstances.
As they should. Why brigade the best, most human game this year? I feel like this drama is blown so out of proportion. Using google and accidentally reading the AI summary level usage.
“everything I don’t like is a brigade!”
Because it’s deceptive. If the end product you’ve built is decent, fun, and polished then there should be no issue being transparent about using AI in your toolkit.
Other games that have been transparent from the beginning aren’t seeing this kind of backlash, because they’re allowing consumers to make informed decisions, and most consumers aren’t waving pitchforks, but rather just don’t want to be bullshitted.
Being honest also will help eventually break the stereotype, that AI toolkit use always results in slop.
and most consumers aren’t waving pitchforks
The readers here are not most consumers and they have several pitchforks.
Posting an article like this is basically saying “Hey, unhinged people of the anti-AI community, go spam the game’s Steam page because they are trying to stop you”
Using google and accidentally reading the AI summary level usage
Using google does not = accidentally reading the ai summary.
Using google means deliberately reading the ai summary.
This community (and it’s sister community on Reddit) do not really have any kind of internally consistent belief structure. The most reductive takes tend to get the most engagement.





