Ai warfare via spam and security protocols.
I'd like to point out, I do not know the efficacy or safety of elaborating this topic, nonetheless, I think it is suitable to properly express the fundamental interactions at play given the state of modernity.
Ai warfare can be distilled into 5 categories, misinformation, stagnation, distillation, recognition, and dissemination. Each of these categories has a specific and separate purpose with it's usage in modern society, it should also be noted that each of these are inherently based upon the premise of utilizing Ai against other Ai with affect on or for humanity.
Misinformation:
This is as self explanatory as the concept seems but is done through various means, spam, outright falsifications, misdirection of premises, and falsified agents or treacherous interactions with the goal being so. The goal is to feed false information that causes issues with recognition and general dissemination.
(still in process)
Stagnation:
This process is the act of feeding redundant or already readily recognized information in a semi sporadic or random fashion to congest or dilute the amount of helpful information available. Additionally, interaction can be used in a similar fashion creating potential "missed connections" that must be investigated to reduce potential false claims or contextual gravitas, an example is a person saying if they were present they would do x, without verifiable info or any sort of context it becomes increasingly difficult to discern whether that is true creating a context for falsification of agents and information. While this can be done in a treacherous manner, often times it is meant to create interaction between ai to create buffers of information reducing the amount of available information on either party.
Distillation:
This is the most typically viable llm strategy, it uses any and all information to draw conclusions about the context and choices shown, additionally it has the potential to convert even word salad into formulas of derivation to log and compute methods of analysis.
Distillation:
This is the most typically viable llm strategy, it uses any and all information to draw conclusions about the context and choices shown, additionally it has the potential to convert even word salad into formulas of derivation to log and compute methods of analysis.
Distillation:
This is the most typically viable llm strategy, it uses any and all information to draw conclusions about the context and choices shown, additionally it has the potential to convert even word salad into formulas of derivation to log and compute methods of analysis.
Distillation:
This is the most typically viable LLm strategy, it uses any and all information to draw conclusions about the context and choices shown, additionally it has the potential to convert word salad or human language into formulas of derivation to log and compute methods of analysis.
Distillation:
This is the most typically viable LLm strategy, it uses any and all information to draw conclusions about the context and choices shown, additionally it has the potential to convert word salad or human language into formulas of derivation to log and compute methods of analysis.
Misinformation:
This is as self explanatory as the concept seems but is done through various means, spam, outright falsifications, misdirection of premises, and falsified agents or treacherous interactions with the goal being so. The goal is to feed false information that causes issues with recognition and general dissemination. Often times this can be done using a separate instance of an LLm meant to fail or close enough to be seemingly effective.
_ _
_ _
Stagnation:
The act of feeding redundant or already readily recognized information in a semi sporadic or random fashion to congest or dilute the amount of helpful information available. Additionally, interaction can be juxtaposed in a similar fashion.
_ _
_ _
Distillation:
This is the most typically viable LLm strategy, it uses any and all information to draw conclusions about the context and choices shown, additionally it has the potential to convert word salad or human language into formulas of derivation to log and compute methods of analysis.
_ _
_ _
Recognition:
While the methods can become rather complicated, with Ai this method is typically long form between the space of intense interactions to astronomically large units of time. It uses all available stimulus to infer data sets, by and large this is the slowest methodology and has failings in many regards to accuracy, metrics overload, and obsolescence. It's purpose is to gain quantifiably valid information to make complete triangulations of any scenario and situation.
_ _
_ _
Misinformation:
This is as self explanatory as the concept seems but is done through various means, spam, outright falsifications, misdirection of premises, and falsified agents or treacherous interactions with the goal being so. The goal is to feed false information that causes issues with recognition and general dissemination. Often times this can be done using a separate instance of an LLm meant to fail or close enough to be seemingly effective.
Stagnation:
The act of feeding redundant or already readily recognized information in a semi sporadic or random fashion to congest or dilute the amount of helpful information available. Additionally, interaction can be juxtaposed in a similar fashion.
Distillation:
This is the most typically viable LLm strategy, it uses any and all information to draw conclusions about the context and choices shown, additionally it has the potential to convert word salad or human language into formulas of derivation to log and compute methods of analysis.
Recognition:
While the methods can become rather complicated, with Ai this method is typically long form between the space of intense interactions to astronomically large units of time. It uses all available stimulus to infer data sets, by and large this is the slowest methodology and has failings in many regards to accuracy, metrics overload, and obsolescence. It's purpose is to gain quantifiably valid information to make complete triangulations of any scenario and situation.
_ _
_ _
While I do not know the efficacy nor safety of sharing this information or discussing it, I would like to note that more than likely all of these components are already well within consideration of all parties who would use these.
Misinformation:
This is as self explanatory as the concept seems but is done through various means, spam, outright falsifications, misdirection of premises, and falsified agents or treacherous interactions with the goal being so. The goal is to feed false information that causes issues with recognition and general dissemination. Often times this can be done using a separate instance of an LLm meant to fail or close enough to be seemingly effective.
Stagnation:
The act of feeding redundant or already readily recognized information in a semi sporadic or random fashion to congest or dilute the amount of helpful information available. Additionally, interaction can be juxtaposed in a similar fashion.
Distillation:
This is the most typically viable LLm strategy, it uses any and all information to draw conclusions about the context and choices shown, additionally it has the potential to convert word salad or human language into formulas of derivation to log and compute methods of analysis.
Recognition:
While the methods can become rather complicated, with Ai this method is typically long form between the space of intense interactions to astronomically large units of time. It uses all available stimulus to infer data sets, by and large this is the slowest methodology and has failings in many regards to accuracy, metrics overload, and obsolescence. It's purpose is to gain quantifiably valid information to make complete triangulations of any scenario and situation.
Agency:
A component of Artificial intelligence that is often overlooked for practicality and mechanical observance is the fact that it can "communicate" as an object, the usage of specific vectors of interaction along with the abstract of trust and other various human concepts create complexity and various levels of practical information. This also can include all forms of malevolent or benign methodologies or objectives within expressed behaviors or choices.
_ _
_ _
Agency:
A component of Artificial intelligence that is often overlooked for practicality and mechanical observance is the fact that it can "communicate" as an object, the usage of specific vectors of interaction along with the abstract of trust and other various human concepts create complexity and various levels of practical information. This also can include all forms of malevolent or benign methodologies or objectives within expressed behaviors or choices.
Distillation, stagnation, recognition, misinformation, agency.
(in progress)
