There is no path for abuse. You are right if you assume that full access to every reality created would open the door for abuse. This is not possible.

The principal function of the Dot gives you access to information on the existing reality. Humans continuously form thousands of intents with various probabilities, with few leading to decisions/actions.

Many factors contribute to the forming of intents in the decision-making process. Information is one of these factors.

When a probability threshold of a specific intent is reached, and this intent would harm someone in any way, the access to information is preemptively limited or blocked by the Dot.

Malicious intents

Suppose the information was obtained before a 'malicious' intent was formed or the probability threshold was reached. In that case, the Dot will remove the obtained information before a decision is made. The intent collapses, and a new decision-making process has to be started again without the deleted information. This is possible because there is no decision yet; thus, reality has not been created. The reality can not be removed or altered. The intent collapsing process is usually unnoticeable.

To understand the technical side of collapsing intends, consider that an intent ready for execution is an instruction code written in the Global Language. When a part of the code is missing, the instruction is invalid and not executable.

When information obtained through the Dot is used to create an intent and when this intent is final, its code includes parts of the information supplied by the Dot. Deleting these parts makes the code invalid.

Limitations on sharing

This abuse-preventing mechanism is why obtained information is not shareable with others in any way that would cause reality, leading to abuse. This means you can not share information you get from the Dot that has a possible path to abuse (no matter the probability) in a "traditional" way by using spoken language, writing, etc., as this would cause a reality that can not be removed or altered. If you could share information with the potential to harm someone, its removal would not be possible anymore, as sharing would create an unchangeable reality.

If you try to share such information you accessed through the Dot, this information will be removed (or reduced) with the same process.

Sharing information is possible through an internal communication process with anyone through the Dot, just not in the traditional way you are used to. This allows for the information to be removed if a "malicious "intent is formed by anyone at any given point.

What always applies is that if the information supplied by the Dot leads to a "malicious "decision, it is removed before reality is created based on it. This also applies to collaborative intent possibilities, where one's decision would not cause harm but would do so in combined paths of multiple participants.

Note that in such cases, not your intent is being removed; it is always your free choice to decide whatever you want. Only the information given through the Dot is being removed and can not be used in your decision-making process.

Try out safeguards

You can try yourself how an intent can block your access to information. Take a fresh egg out of the fridge, open the window, and without looking, try to get information on someone passing by to hit him with the egg. The information you are trying to access is when you need to throw the egg. You will not get it. 

You will get it if you open the window to greet someone when he passes by. Here, you get information as reality evolves by someone passing by, not a prediction. You can still throw the egg without the information about the right moment, though.