AntiRAID.AI uses ChatGPT AI to analyze each message to determine if it contains spam, solicitation, FUD, or requests to check DMs.
Scans the group for user-defined block phrases and banned member text to detect semantic similarity between previous undesirable messages and current messages.
There are multiple sensitivity levels for the bot that allow admins to control how similar current messages must be to banned text before they are considered a match.
Bio scanner scans new users bios and detects spam and bot accounts on sight.
Ask questions and interact with the AntiRAID Community
Stay up to date with the latest news from AntiRAID
Displays the welcome message and a brief introduction to what I do.
Displays the help section you are reading.
Use this by replying to another message, blocks the user and saves the message into a blocklist to compare future messages in chat against.
bans the last user who joined your group
un-bans the last user that was banned by the bot, in case they are banned by mistake. In a couple of days we will be giving you the option to kick, warn or ban and in later updates have custom settings for different lists of tags.
sets the name of what the group/product/coin is about as it is submitted to OpenAI when making queries about new messages. By default it is the same as the title of your group but you may wish for it to be shorter/different for better results.
same as setcontext_name but for the description of the group/product/coin. Again, by default it is the same as your group's description. You may wish to condense or better summarize what your group is about for better results.
displays a list of the current blockphrases in your group.
deletes the blockphrase corresponding to the given index number. To check the index number of all items on the blocklist use the command /blockphrases
wipes all block phrases currently associated with your group.
chooses from 5 levels of sensitivity for embeddings comparisons, where 1 is the least sensitive (least likely to trigger a match) and 5 is the most sensitive (most likely to trigger a match, with higher risk of false positives).
chooses whether to mute or ban the user when they have exhausted all warnings and a match has been triggered.
specifies a number of warnings to be given to the user upon matches being triggered before they are muted/banned.
shows the current settings for AntiRAID.AI for your specific group.
Launch token on Uniswap and release AntiRAID.AI to the public. Initiate shilling and advertising campaigns on Twitter, 4chan, coin listing sites and obtain Buy Bot Tech Trending.
The bot will be upgraded to allow each message to be queried by ChatGPT to analyze whether it contains spam, solicitation, FUD or requests to check DM - vastly improving its ability to detect unwanted messages. Admins will be able to adjust whether the bot mutes or bans, specify a number of warnings to be given, adjust the sensitivity of the bot over 5 levels, and remove individual items from the blocklist. The bot will also gain the ability to read user bios and automatically ban marketers with certain key words in their bio.
Add smart raid detection feature which measures the similarity between posts from new users and triggers 'RAID MODE' if a certain number of similar messages are detected. Offending messages automatically added to blocklist and sensitivity temporarily increased. At this stage we will add an official AntiRAID.AI Portal that utilizes CAPTCHAs on a public channel available to any group to protect against auto-joining spam bots. After passing the CAPTCHA and entering the group, the admins will also have the option to require new members to rephrase a simple sentence to prove they are human or be removed from the group.
Train GPT-3 specifically with text from crypto Telegram groups to allow it to understand the semantic relationship across various internet slang terminology (premium feature). Integrate the $MOD.AI() token into the product by requiring admins to hold a specified number of tokens in their wallet to maintain access to AntiRAID Premium. Add a 'global blocklist' which uses AI to scan messages added to the blocklist in multiple groups and adds any messages or phrases similar or identical in more than one group to the 'global blocklist' in every AntiRAID Premium subscribing group. When tokenization is added to the bot, we will also require token holding for use of the ChatGPT analysis as a premium feature.
Add all general features found in community management bots such as Rose, MEVFree Guardian, SHIELDY etc into AntiRAID.AI such as greetings, filters etc. AntiRAID.AI will also be programmed to randomly join in conversations using GPT-3's text generator at this stage. Add support for Google's LaMDA AI for AntiRAID Premium subscribers and allow it to be used as an alternative to or in combination with GPT-3/GPT-4 to detect spam. When released by OpenAI, we will also include more advanced embeddings models with up to 8x the number of dimensions for more powerful blocklist comparisons. A Discord bot with all of the above features will be released at this stage. With the release of GPT-4, we will be offering use of this language model for analysis of the chat for token holders only. GPT-4 is expected to have dramatically more parameters than GPT-3 and have a much greater understanding of context and nuance, making the bot even more intelligent and powerful as a community moderation tool.
AntiRaid.AI
Copyright 2023