Discord AI Problems

Every platform has its malfunctions at times, some are a little more severe than others, some even getting your account banned for a year. Sadly, these malfunctions have been popping up more frequently recently and the causes all of it can be easily pointed out, at least if all accusations are true. 

Unrightful banning of accounts

Getting your account banned by Discord for a certain amount of time may seem too far reached, but people have been experiencing it more and more recently. Before accusations are thrown that they did do something wrong, or else you wouldn't get your account banned, but they didn't. You can get banned even as simple as logging into your account or sending the text "me 2".

Thanks to multiple Reddit communities, people can share these observations with each other and, hopefully, discover the reason behind this happening. Yes, this info comes from Reddit because it is impossible to investigate this ourselves. After all, we don't exactly know what causes it and wouldn't want to risk getting more accounts banned for the wrong reasons. By collecting all posts, going from posted 2 hours ago to even a month, some clear wrongful accusations have been collected.

One of these is using numbers in messages as already mentioned before. A simple 2 is already enough to cause your account to get banned for breaking the minimum age requirement, which is 13 as listed in the ToS. This can not be the work of a human but of a bot, specifically an AI bot. 

Analysing messages as an AI

With the surprise of AI in the world in the last few years, more and more companies have taken AI under their wing to minimise the workload on their human employees. Yet, sadly, these AI are not as trained or experienced as the human mind would be, which makes it impossible for AI to make every decision as we humans do.

If you analyse the message "me 2" as an AI, and you must think a dumb way about this, you can quickly misinterpret the message. People love to shorten their messages, making it easier to send them quick so the conversation isn't broken or their message is sent too late. If you take this in the context of telling your age to someone in a convo where the previous messages were "I'm 18 years old" followed by "me 16", reacting with the message "me 2" as saying "I'm 16 too" (2 used as too) could be misinterpreted as "I'm 2 years old".

As experienced I am with AI, AI does intend not to read its or the user's past messages if it's not trained to do so, creating a confusing sphere such as in this case. 

More AI in Discord

Even more, AI has been discovered not only within Discord but also in the support emails you sent to Discord are replied to by an AI. Luckily, these conversations can be a little more fun than they should be. And, of course, as creative as Discord users can be, one of those bots has now become a cooking expert in some way.

I'm talking about Nelly, one of the Discord support bots you email too when you have a problem with something in Discord. You tell you're a problem to the bot, and they reply back asking to confirm the situation before sending it on to whoever knows where. Yet when you correct the situation, the bot forgets the previous email completely, which can be very funny to use. If you then, for example, ask to tell you a perfect recipe for spaghetti, it serves it to you like a food critic. 

The bad side to all of this

If AI is already monitoring the support emails and flagging or banning accounts, you can't imagine what more problems there might be due to the whole AI integration. Problems within servers are left untouched, giving access to bad organisations to take down other servers, accounts and whole organisations. One of these possible outcomes is explained in the last post, showing that Discord is leaving harmful bots within its platform without doing anything about it or the AI picking them up. 

Protect your account!!

I personally advise you to enable 2FA in Discord, needing to have a code that you can access through an app on your phone to secure your account. Remove any info on your account that could possibly harm you in your personal life in case your account meets unrightful banning or worse things could happen to it such as your phone number and your email address. Don't connect too many social media to your account, disable that people can see what you play and log out on your device once you won't use it on that device anymore. Don't interact with applications you don't know, don't click on links that seem suspicious, and don't authorise bots to your account. 

You can't completely protect your account from these actions or other bad intentions to the point others won't be able to get into your account or do anything with it. The least you can do is minimalise the threat.