Good LLMs that handle bias - Introducing IAN

If you want to find a truly unbiased AI tool, come try out IAN.

Here is the link for those that worry about this stuff:

https://mee6.xyz/i/fA2U1AaiVg

This will get you into the community and help you understand how IAN is making unbiased LLM a reality. They use multiple LLM models to compare answers and provide the most unbiased answer possible. The models chosen are evaluated for their answers.

It will be available soon. Come see how you can participate in the community. They are hoping that their IAN tokens will soon turn into cash through crypto. I was building a system like this inside of the Diassu Safe Identity, but now I might change courses and use their system. Why re-invent the wheel at Diassu Software?

I have been working with an amazing group of folks on this lately, and they plan to launch soon.

Not like the other LLMs (BING etc.) built by liberals that won't answer the simple question, claiming that it is politically related. Why do you not want to answer some questions? LLMs should answer all questions and say, hey, this is based upon public knowledge that might be wrong.

The following NON POLITICAL QUESTION, is a sincere search for a job and some pay to support my family.

I truly love Bing Chat and what it does for me each day, so please don't get the wrong impression about what I am saying, but look at what happened above when I asked the following question to Bing Chat:

How do I find a Job in the new Donald Trump Administration?

As you can see, Bing below thinks it is a political question:

Even wasting 2 minutes of my time right now is costing me food on my table. Liberals drive me nuts with this kind of waste of time coding and they create programs that waste time as well.

I want AI to help me find a job.... Not stop me from finding a job. When I asked IAN about what the current LLM models thought about my experience, they gave poor results at best. This means that they are all biased against me. How is that helping me? I have 35 years and graduated top of my class in Computer Science, so you can kind of trust what I am saying: I hope?

We need to defund these types of companies and groups that are against the American way and that find any excuse they can to BAN your voice and your First Amendment Rights. We need to stop building LLM models that go against us. We need to identify tools to make them stay honest. I think IAN is one of those tools.

Censoring, Banning, Filtering, Misinformation all ring of limiting your First Amendment.

A code of conduct is different than what I mean here. Healthy debate is good. Someone that disagrees with you is a good thing sometimes. Someone who blasts it to you is good as well. Learn to love your enemy if you can. I am trying to while at the same time expressing my opinion.

What about LLM Bias though? LLM's are biased based upon the information you train them with, but should we balance the Democrat articles that we train our LLMs with against the Republican articles that we train them with? Should we use the same number of Chinese, Japanese, black people photograph as white people etc. etc. and the list goes on?

I say no and that we should not care that much because we are wasting our time on things that don't really matter that much in the real world. DEI for example just allows us to pick a less than qualified candidate because they fit into a category. I have been replaced 3 times in my career this way and 2 of the 3 times I was more qualified.

Let's pick the people with the skills we need and stop thinking that we need some system that can be rigged. If you build it, they will rig it I say.

If you train on all of the data, you will have at least all of the ground truth. We should not care as much as we do because this is where the Liberal Media wants you to believe and then there is the ground truth.

Why not train the LLM on everything on the internet and then if it is biased that month, that is because more of one party spoke than the other. Why did Google have the problems they did? They did not train the LLM on our founding father's images? How could you leave that out? That is because you don't want the founding fathers to be white people? It brings this to mind in my garbage fascist mind. Not really! I am exaggerating here to show you how easy it is to misunderstand something.

I think a vocal America is a healthy America because when you repress your thoughts that leads to something called psychological repression and then you need anti-depressants. We should debate these things, and I welcome a debate below in the comments. Do you know that some anti-depressants can cause suicide? Surely take them if you need them but carefully weigh what is worse.

Imagine if you tell a terrorist that you don't want to tell him how to build a bomb because it is bad for society, but the person asking is a guy in a ditch in the Ukraine? Imagine if he is an 18-year-old that needs help, and you did not let someone know he was possibly in trouble with a question like that? Imagine a IT manager with little experience being sold to replace a group of help desk agents with an AI that is only right 65% of the time. Big trouble on the road ahead or are we already the one with the smashed back end?