DeutschEnglish
Nicht eingeloggt oder registriert. | Einloggen | Registrieren | Passwort vergessen?

Website Updates: Clarification to ratings system rules regarding AI-generated comments

WebsiteUpdates

Artikeldaten
Veröffentlicht am:
06.08.2023, 07:40 Uhr
Kategorie:
Serie:
CAGEMATCH Website Updates (Alle Folgen der Serie auf einen Blick)
Verfasser:
Hello everyone!

First of, I hope you enjoyed your dose of wrestling over the last few days, regardless of what company it may be from.

That being said, some members of the community have been exposed to using large language models or other "artificial intelligence" tools (example: ChatGPT) for their comments recently. I noticed this myself when looking over the SummerSlam reviews from last night, but thanks to some observant users more examples where given as well. I will take action today and remove those users from the ratings system.

In my personal opinion, the rules of the ratings system were already clear insofar that written comments must be from you and not another entitity. While technically the verbiage in the rules uses "another user or website", the intention should have been clear that AI-generated comments are considered detrimental to the community. Just to make this particularly clear, I added a new bullet point to chapter 2.1 of the ratings system rules just now that specifically mentions artificial intelligence.

Look, I get it, artificial intelligence is a hot topic. At my real-life workplace I am involved in a project that evaluates its potential use in our workplace environment in regards to benefits and usability as an assistance tool. More generally speaking, I believe it will be a great challenge for us as a society to find out how to use AI as constructively as possible without doing more harm than good in the process. Naturally, the range of acceptance, motivations and understanding by actual human beings will be (and already is) all over the place. It makes sense that this is also true for the inmates of this community.

But, what the AI afficionados in our midst must understand is that AI-generated text is not welcome content for our ratings system and it is most certainly not helpful or interesting for most users to read. Here is an example from last night's WWE SummerSlam (without spoilers), which is quite obvious AI-generated text and adds zero value:

"Recognizing SummerSlam's status as one of WWE's premier events, it is traditionally granted a longer runtime to emphasize its significance. However, the argument remains that the five-hour duration (including the Kickoff Show) may have exceeded the optimal threshold for a wrestling spectacle. By condensing the event, WWE could have maintained a consistently engaging experience, catering to both the live crowd and the viewers at home without compromising the show's grandeur. I don't think there were any bad matches on the show."

Naturally, in a few minutes you will not find this review on the show anymore because the user will have been warned (or banned, if repeat offender or other circumstances apply).

I would also like to use this opportunity to thank those of you who help with identifying bad faith actors of the ratings system, very specifically in this case those who help identifying AI-generated comments. Using an "AI content detector" like CopyLeaks is not something that is easy for us to do, as these tools only allow a few checks per day per user unless you pay an obscene amount of money. With the amount of comments we get per day it would cost a few hundred euros per month to check all of them, which frankly is not something I am willing to do -- and I also would like to keep this website free to use. Therefore I am extra thankful for those of you who help identifying the AI-generated comments by using your "daily free uses" of such tools for suspicious comments on our site. Thank you.

That's it. You can now resume enjoying wrestling of any company that you choose.

Have a nice day!