Bot Posts

Can we not implement a system to slow down the bot posting? At least not allow new accounts to post in the first day, first week or whatever. This is ridiculous. 

I’ve posted about his before, recently. I think that people who call themselves “moderators“ should be checking in every day

LL2, I think there are simpler ways. If bots can spam these type of posts, a system should easily be able to detect them. 10 nearly identical posts in an hour seems very simple to detect.

I would agree Walden. A potential approach for this would be generating vector embeddings for each new post (title + first post contents). This post could be compared to other posts by the user within a recent time frame. If they have a high cosine similarity then they are likely a spam post. This could be applied for newer accounts until they reach a certain threshold which shows organic user activity. 

The cost for this would be very minimal as text embeddings typically bill at 0.02 per 1M input tokens. A simple formula for this is:

Cost = ( new accounts per day ) • (tokens per post) •  (.02/1M)

(10,000) • (420 ) • (.02/1M)

= $0.084 (obviously this assumes 1 post per bot, but I'm overestimating on new user signups probably)

False positives could be reduced by 

- Requiring more than X posts to be in violation.

- Checking semantic similarity with typical rock climbing posts. Ie. Is it about rock climbing at all? 

- Using TF/IDF as an embedding component for the site. We can better represent/factor in each sub-fourm's discussion nuances. 

- Comparing the embeddings of a user to known past spammers.

Idk, I don't work in this space and I'm probably over simplifying/not factoring in a lot of nuances, but vector embeddings are sweet and they seem well suited to typical spam posters I've seen in the past few months.

We're working on making improvements to the current system to prevent or limit similar bot-spam situations in the future. In these cases, users were quick to flag the posts but we don't have an automated system for deleting posts. That has been an intentional choice to limit the possibility of making mistakes and removing legit user posts. It's all a balance and the feedback is appreciated. 

Most of the accounts that post on day 1 are bots. Most humans who are new to MP and post immediately are asking questions that could be answered with a simple search on MP or even Google. Such users have more to gain by being forced to do their own research than being allowed to ask dumb/lazy questions.

The only compelling reason to allow a brand new account to post is an emergency situation. I.e. someone is trying to help search parties find missing person and is hoping one of us has a clue, or a dangerous person is creeping around Maple Canyon.

My suggestions: 

• New users are barred from posting in most forums. Exceptions might be General Climbing, Regional Forums, Women's , and Accidents. 
• When a new user attempts to post, they get a pop up telling them that they are only allowed to create posts related to emergency situations and if they violate this rule, their account will be deleted. To post, they must check a box stating that it's an emergency. Perhaps a second Turing test is added here.
• Such new posts automatically generate a message to Moderators. If it's obviously not an emergency, moderator deletes the account with a couple of clicks.

As for the duration of new user restrictions, I'd defer to someone with a better understanding of AI. But I don't think a week or even a month is draconian. One can enjoy 90+% of MP's value without ever creating a post. I realize that smarter bots might have attention spans, but would hope that this protocol would at least flag a high percentage of bots before they can post.