An AI Guide To AI

Can we prevent AI from being used against us and of abusing the confidence we allow it?

As Snowden pointed out a few weeks ago, the formidable power of AI can be tainted with lack of transparency in the way it trains itself when interacting with users’ data and what it stores and deletes.

As we found out in the Cambridge Analytica debacle, user data has become a very valuable commodity and the high degree of intimacy that is built between AI and human users can create a huge repository of detailed qualitative data about millions of individuals with an unprecedented capacity to manipulate the masses.

Another major issue with LLM is the question of bias. Examples such as images depicting German Nazi soldiers as ethnic minorities, illustrated the absurdity auto censorship can bring about. Training in political correctness, though maybe used for the right reasons, can lead to unacceptable censorship and reverse bias resulting in misinformation.

Learn more about AI bias in An AI Guide To Ai

In an article on cointelegraph.com, Andrew Fenton has published a detailed guide to the censorship, bias and anonymity safeguards found in major LLMs and has proposed more neutral alternatives to the more mainstream offerings.

Cypherpunk AI: Guide to uncensored, unbiased, anonymous AI in 2025

“As the experience of Elon Musk buying Twitter shows, the political orientation of centralized platforms can flip on a dime. That means both the left and the right — perhaps even democracy itself — are at risk from biased AI models controlled by a handful of powerful corporations. ”
Andrew Fenton

https://cointelegraph.com/magazine/cypherpunk-ai-guide-to-uncensored-unbiased-anonymous-ai-in-2025/

Fenton reviews uncensored AI models: Dolphin Llama, Dophin Mistral, Flux Custom and private AI platforms: Venice.ai, Duck.ai which can be used to privately access larger models in private and discusses their pros and cons.

Learning about these alternative models allows you to work in a more transparent manner if you don’t fully trust the offerings of the wealthiest people of the planet (Gates, Bezos, Musk and Zuckerberg) to not abuse of the power this knowledge bestows on them.

Learn how to use AI in a safer and less manipulative environment.