We are in an era where mechanisms of mind manipulation work on most of us silently, in the form of creeping gaslighting using powerful technological tools — in the years 2015-2025, for example, a quiet coup d'état took place in Israel based on the spread of disinformation, distractions, and distortion of facts. In the current public discourse, people find themselves trapped in "mass hypnosis" — a situation in which it is difficult to distinguish between truth and lies, who is behind the messages, and what the real motives are.
Whether I am right in this statement or not, this tool will be able to provide the factual support.
The goal of this project is to give every person a means of self-defense — to understand who is really running the public discourse, to identify when "mind operations" are being carried out on them such as Netanyau excecuted with the German Bild, and to extract from the chaos the real and reliable facts. This tool is intended to appeal to the entire world, not just Israelis, but with awareness of the unique facts of Israel and for which it is trained — among other things, a quiet coup that occurred in the last decade through gaslighting mechanisms and the creation of a deep rift between different worldviews.
Once we identify the manipulations and understand the facts, public discourse will be able to move to a new and deeper stage: an open, conscious, and authentic discussion of the fundamental principles that make up society.
In Israel, for example, the discourse needs to deal with the deep gap between two distinct worldviews — a state with a religious, messianic Jewish identity versus a liberal, secular state. These are two value worlds that do not integrate easily or at all, and society must consciously decide how to deal with this gap: whether by agreement, separation of populations, open struggle, or any other solution, to be decided by the public itself — and not as a tool of powerful elements who govern us from afar.
The YARRIVA_bot project is not just a tool for detecting manipulations and extracting facts — but a platform that aims to enable real, transparent, and rational discourse. It allows every person to take control of their own consciousness, develop critical thinking, and avoid being a player in a "chess game" of those with wealth and power, as was done, for example, by Cambridge Analytica.
Why Twitter? Not because this is the ultimate goal, but because it is currently the most suitable platform for building the training files for the AI behind the tool, and this is the heart of the project (the trade secret). The system itself is flexible and can be implemented in any other environment in the future — so that the tool can be expanded to any media that the AI engine has access to and to any community in the world, including places that have undergone similar processes, such as Hungary, and many others that face the same challenge of a quiet coup and restrictions on freedom of information.
The vision is that every person will be able to make informed and stable decisions, based on clear facts and values, and not be dependent on external manipulations aimed at controlling and dividing society into opposing and controlled forces and unnecessary wars.
How to Use
YARRIVA_bot is designed to help you analyze posts that you feel are manipulative, misleading, or deceptive, and uncover the truth behind them. It is simple and intuitive to use, allowing you to take control of your public discourse in a well-founded and factual manner.
Here are the steps to activate the bot:
Tag the bot in a response to the post on X.com that you want to analyze.
Select a Twitter post that you suspect is manipulated, gaslighted, or makes you feel uncomfortable, but you have difficulty pinpointing exactly what the problem is. Tag @YarrivaCom in a response to this post, so the bot can identify it and start processing the content.
Choose the appropriate command for the action you want to perform
The bot will post its analysis on its own wall.
There are two main commands you can use:
CutTheBS — This command asks the bot to expose the mechanisms of manipulation, deception, and gaslighting in the text, taking into account the comments and contexts you add. This is the starting point for understanding what is behind the problematic messages.
GetTheClaims — This command allows you to extract the main claims and precise facts from the text, based on the initial analysis performed in CutTheBS (even if you did not call this command directly, so it is supposedly "more expensive"). This is the way to summarize the objective information that can be extracted from the text.
Wait for the result.
The bot performs advanced analysis of the text depending on the existing request load. The waiting time to receive the answer varies depending on the bot's activity on the network. It is worth being patient and checking the results on the bot's Twitter page. When the processing is complete, the bot will respond to your response.
Use the results for respectful, relevant discourse and dissemination of information.
After receiving the analysis, it is recommended to publish a response to the original post with the insights and conclusions you received from the bot. This will help promote a transparent, rational, and fact-based discussion, and prevent further dissemination of misleading or manipulative information. Also, sharing the results supports the dissemination of the idea and the tool itself, and allows for the building of an active community around the bot.
Important to know:
Currently, the bot is open and free to use in the initial development phase.
I invite you to send suggestions, comments, and ideas that will improve the tool and enrich its capabilities.
In the future, additional commands and tools will be added that will contribute to creating a platform for managing social discourse more efficiently, transparently, and democratically.
MVP Phase: Completed
The two main commands are active. The bot responds to tags, detects manipulations, summarizes claims, and also addresses attached images. At this stage, the improvement process continues manually through additional analyses fed into the training file.
Possible future phases:
Establishment Phase
The intermediate products of the analyses will be opened for public viewing. The purpose of the opening is to enable transparent review, monitoring, and identification of failures. Each analysis constitutes a link in a chain of cumulative processes – similar to a network of bots that use the products of their predecessors.
Expansion Phase
Support for video analysis is planned, similar to the already active image analysis. Additional commands will be added for rhetoric analysis, pattern recognition, internal contradictions, or inconsistent claims.
At this stage, the integration of community contributions into the training process will also be examined – in a centralized, non-automatic manner.
Structural phase
A graphical interface is planned for presenting claims and problems in a hierarchical structure, to enable understanding, navigation and in-depth analysis by topic. The interface will incorporate the possibility of contributing to the public debate, while creating an incremental solution tree.
Systemic phase
Long-term planning includes a broad mapping of the public discourse in a way that will neutralize manipulations, while breaking down claims into testable components. Mechanisms will be examined to unify duplicate claims and build a platform for presenting facts, links and sources, which will enable scientific discourse to be conducted as much as possible – based on claims that are subject to testing, context and references to sources.
