AI as Staff/Moderators?
- Thread starter Katrina
- Start date
To all intents and purposes the UK's Online Safety Act (OSA) makes that mandatory both here in the UK and arguably abroad.
It's late here and I don't currently have time to go into it in any great detail but there's a long thread which explains more on Xenforo's customer site.
UK Online Safety Regulations and impact on Forums
The UK regulator OFCOM yesterday announced a set of regulations which will come into effect on 17th March 2025 https://www.ofcom.org.uk/online-safety/illegal-and-harmful-content/time-for-tech-firms-to-act-uk-online-safety-regulation-comes-into-force/ These have broad implications for all...
xenforo.com
In short OSA requires all site owners to conduct a detail risk assessment with a view to implementing AI driven software in order to carry out numerous tasks which might include robust age verification and content scanning to ensure only age appropriate content.
Some further reading:
Online Safety Act: explainer
First UK-US online safety agreement pledges closer co-operation to keep children safe online
Statement between the UK and US will bring countries closer on joint priority of creating a safer online world.
First Ofcom probe launched into suicide site exposed by BBC
BBC News found the forum, which has thousands of members, is linked to at least 50 deaths in the UK.
www.bbc.co.uk
While I have some sympathy with organizations charged with keeping children safe from harm, I personally don't entirely agree with the method being employed here for more than one reason.
Death Of A Forum: How The UK’s Online Safety Act Is Killing Communities
We’ve been warning for years that the UK’s Online Safety Act would be a disaster for the open internet. Its supporters accused us of exaggerating, or “shilling” for Big Tech…
www.techdirt.com
I'll probably have more to say on this later.
To all intents and purposes the UK's Online Safety Act (OSA) makes that mandatory both here in the UK and arguably abroad.
It's late here and I don't currently have time to go into it in any great detail but there's a long thread which explains more on Xenforo's customer site.
UK Online Safety Regulations and impact on Forums
The UK regulator OFCOM yesterday announced a set of regulations which will come into effect on 17th March 2025 https://www.ofcom.org.uk/online-safety/illegal-and-harmful-content/time-for-tech-firms-to-act-uk-online-safety-regulation-comes-into-force/ These have broad implications for all...
In short OSA requires all site owners to conduct a detail risk assessment with a view to implementing AI driven software in order to carry out numerous tasks which might include robust age verification and content scanning to ensure only age appropriate content.
Some further reading:
Online Safety Act: explainer
www.gov.uk
First UK-US online safety agreement pledges closer co-operation to keep children safe online
Statement between the UK and US will bring countries closer on joint priority of creating a safer online world.
First Ofcom probe launched into suicide site exposed by BBC
BBC News found the forum, which has thousands of members, is linked to at least 50 deaths in the UK.
While I have some sympathy with organizations charged with keeping children safe from harm, I personally don't entirely agree with the method being employed here for more than one reason.
Death Of A Forum: How The UK’s Online Safety Act Is Killing Communities
We’ve been warning for years that the UK’s Online Safety Act would be a disaster for the open internet. Its supporters accused us of exaggerating, or “shilling” for Big Tech…
I'll probably have more to say on this later.
How about having AI give a warning to a member who commits a rule violation?
As long as it's transparent.
I have a Facebook group with about 50K members. It gets a lot of activity and new members daily.
Facebook's AdminAssist is powered by AI. We also have an AI chat bot. We're very transparent about it. The AI is being used to help us catch spam and abuse before Facebook removes it. We don't use it to post new content to the group.
I feel quite strongly that members need to be treated as individuals, something AI is currently not yet capable of. I'd certainly use AI to alert me or other staff to bad behaviour but I probably wouldn't use it to issue a warning directly to a member. I might think differently if I was running a forum with a very high daily post count though.
I can see an AI assistant on a resource forum being helpful for when the humans are not online. That way if members in a different country need help an AI assistant could fill in until a human can. Other than that I don't see how AI moderation could be a good thing. I see automods on subreddits remove comments for no reason, only for human moderators to reapprove comments.
I think it's rather unfortunate that the focus of AI in many areas including forums is as a content creator rather than a help and support tool.
From a member point of view I'd like to see AI used as a universal translator and to provide optional assistance with spelling, formatting and grammar. It could also help with predictive search/similar topics and image search.
For the a owner/moderator; better seo, robust age verification, text and image search for content not deemed age appropriate especially within private messaging. Early warnings for inappropriate content. Better handling of spam accounts.
From a member point of view I'd like to see AI used as a universal translator and to provide optional assistance with spelling, formatting and grammar. It could also help with predictive search/similar topics and image search.
For the a owner/moderator; better seo, robust age verification, text and image search for content not deemed age appropriate especially within private messaging. Early warnings for inappropriate content. Better handling of spam accounts.
Using it for SEO is one of the key things that it's good for, especially for semantic seo and AI overviews. You can have it use your current text, improve it and enhance it just for those core fundametials, which is exactly what AI(a tool) was designed for.
It was never created to be more than just a tool, however, some are using it to their advantage and making fluff across the internet, which is quite sad & disappointing.
A script was recently developed where you can use AI as a moderation tool as well, but it requires Next.js, MongoDB, Permit.io, and Gemini 1.5 Flash. It may be possible to get it ported over to Xenforo and various other softwares though.
I think it could be quite beneficial as a tool to help with moderation, as others have said. If no one's available in that moment, AI could help moderate the content and allow it to go under review before it becomes available to the public. That way, when a human is available to review it, they can make sure it falls within community guidelines before allowing the post to go live. It'd allow for it to be tweaked too to make it better with moderating things on the forum and not flag things unnecessarily. That said, I think it's important to make users aware that AI moderation is active on the site to be transparent and in case things are accidentally flagged for no reason.
For content creation though, I feel like staff on a forum would be at least moderately knowledgeable enough to create content on their forum without the need for AI, or so I would hope. I could see it being useful to help get things lively in areas of the site that aren't so active that you may not be as knowledgeable on, but it would also be worth editing to make it sound less like AI and more like a human created the post.
For content creation though, I feel like staff on a forum would be at least moderately knowledgeable enough to create content on their forum without the need for AI, or so I would hope. I could see it being useful to help get things lively in areas of the site that aren't so active that you may not be as knowledgeable on, but it would also be worth editing to make it sound less like AI and more like a human created the post.
You wish!
I am totally against having AI in forums. The very philosophy of forums is a platform for people to get together, not computers to come together.
AI in any roles, members/posters/moderators is just not my cup of tea.
AI in any roles, members/posters/moderators is just not my cup of tea.
Kennysplash
Acquaintance
I visited a forum where AI was doing everything. It created threads and give replies to every comment under the thread.
It was an eyesore.
I don’t support the idea AI, especially as a content creator (or moderator even.)
It was an eyesore.
I don’t support the idea AI, especially as a content creator (or moderator even.)
Hopefully they see that it wasn't helping the forum as they intended? It can really turn people off.
Affiliates
