Forum Bots

I'd recently been hearing of this happening more often: bots posing as human users on a discussion forum.
I asked an AI platform about this and sharing what came out. I will be keeping this in mind on the discussion boards that I participate in:

"Bots can and frequently do pose as real users on forums and other online platforms. With advances in AI, these "social bots" are increasingly sophisticated and can be very difficult to distinguish from genuine human accounts.
In a 2025 experiment, researchers deployed a series of AI bots on Reddit that posed as real people and engaged with users to try to influence their opinions. This demonstrated the practical capability of bots to integrate into a forum environment.
How bots impersonate humans on forums
Mimicking human language: Modern bots, especially those powered by large language models (LLMs), can produce coherent, context-aware, and natural-sounding text that is almost indistinguishable from a human's. They can mirror tone and language patterns to seem more authentic.
Creating fake identities: Bots can create convincing profiles with generic or stolen usernames and profile pictures. To build credibility, they may participate in seemingly legitimate conversations for months or even years before being used for a malicious purpose.
Copying and amplifying content: Reposting is a common bot tactic. They copy and paste popular comments from past threads or plagiarize content from real users to build up a posting history. This activity also amplifies specific messages, creating the illusion of consensus or widespread support.
Simulating realistic behavior: Some advanced bots can mimic human-like inconsistencies, such as typos, variable posting times, or engagement with different topics. This helps them evade detection by behavioral analysis tools.
How to spot a bot on a forum
While modern bots are sophisticated, they still often exhibit certain behaviors that can betray their true nature:
Unusual posting activity: A profile that posts at an unnaturally high frequency or at odd hours is suspicious. A high volume of identical or near-identical posts across different threads is also a major red flag.
Lack of depth in conversation: Bots often struggle with complex, nuanced, or off-script conversations. They may give generic or repetitive responses, redirect the topic, or ignore specific questions.
Generic or stolen profile information: Scrutinize usernames that are a string of unrelated words and numbers. Perform a reverse image search on profile pictures to see if they appear elsewhere on the internet.
Focus on a single, narrow topic: A bot may focus all of its energy on promoting a product or a specific political viewpoint. It will lack the diverse interests typical of a human user.
Suspicious links: Be wary of profiles that consistently post links to external websites, especially if the links are unrelated to the discussion or promote scams.
Account age and karma: Check the account's history. A relatively new account with a high karma score (on sites like Reddit) that posts or comments aggressively can be a bot.

Create an Account or Log In

Membership is free. Simply choose your username, type in your email address, and choose a password. You immediately get full access to the forum.

Already a member? Log In.

Very weak and questionable in place security for those with unwanted bots. Great read. Some of these help and chat bots' responses are hilarious when you try to solicit more information from a business.
My experiences continue to be unsatisfying, if not laughable.

In the process of updating all my music software on a new-to-me laptop in advance of my presentations for school, and one company doesn't have the codes displayed with the purchase history info. Their support bot does not like being told its advice is incorrect.

Companies relying too heavily on non-human support may wind up saving money, but it will cost them in customer loyalty in the long run.

Have synthesizers, will travel...
USAA, my insurance company and bank appears to be using a bot on their chat. I have yet to find it useful and end up calling them.

Edited 1 time(s). Last edit at 08/27/2025 09:29AM by wrosie.
I made a reservation yesterday for Amtrak, using the chat feature on their site.
I am pretty good at detecting bots. If that was a bot helping me, they were very good.
Which would be very surprising for Amtrak.
Reddit has issues with bots posing as regular posters, so I have been on the lookout here and elsewhere....some of the tipoffs are uncanny - as well, security issues are a constant cat and mouse game.

Edited 1 time(s). Last edit at 08/27/2025 09:54AM by BarefootBliss.
The only company that I know of that has live people in their customer service chats, seems to be Amazon, and you still have to convince the AI bot that it's something they can't help with first. Every other customer service chat seems to use bots, Verizon, T-Mobile, even the banks that I deal with. Every one of them makes you run through a standard check list before they will let you speak to a human being who can actually solve your problem.
@CoolMusic wrote:

My experiences continue to be unsatisfying, if not laughable.

In the process of updating all my music software on a new-to-me laptop in advance of my presentations for school, and one company doesn't have the codes displayed with the purchase history info. Their support bot does not like being told its advice is incorrect.

Companies relying too heavily on non-human support may wind up saving money, but it will cost them in customer loyalty in the long run.

Great topic Barefoot.

I provide this feedback with factual evidence as often as I can. Most company decision makers are all bottom liners who could care less until it bites them in thier …bottom line.

Most AI isn’t even tested by the decision makers as though they were prospective consumers to experience the chaos and frustration of a poorly programmed bot first hand.

Meanwhile the real people on the front lines (who wish AI really was assistive) have to live with and sort out the dystopian dysfunction are often blamed and possibly terminated due to reduced overall corporate earnings.

…something to keep in mind as we do our evaluations.

Edited 2 time(s). Last edit at 08/28/2025 02:01AM by SBP.
The Dead Internet Theory is quite old.

shopping north west PA and south west ny
I know what DIT is, but can you clarify your point a bit further?
a signifcant portion of the internet is just bots talking to each other. i remeber how bad yahoochat was with bots25 years ago. now its AI which are way more powerful bots. they are currently posting triggering content to control the narrative and thought process of humans. reddit or facebook are now full of them. its exceedingly harder and harder to tell

shopping north west PA and south west ny
Sorry, only registered users may post in this forum.

Click here to login