Help Making most logical and rational thinking AI

Beginners
1

i will keep this short

  • I’ve spent weeks studying and fine-tuning my understanding of AI.
  • I want to create an AI focused on logical and rational thinking, no matter the topic.
  • The goal is to handle conflicting information by using thousands of books on a wide range of topics.
  • Many books provide both sides of an argument, explaining why each can be right or wrong.
  • I aim to train the AI to navigate these complexities and make reasoned decisions.

I need help with the following areas:

  1. Code for AI development:
  • What programming languages and libraries would you recommend for building an AI that focuses on logical and rational thinking?
  1. Data extraction:
  • I’ve worked with JSON (both complex and simple) but need help with effective data extraction.
  • I’ve tried using NLP for data extraction, but the results are noisy, even after applying many custom rules (like removing spaces).
  • What methods or tools can I use to improve the extraction process from the books or data sources I’m working with?

so i though to build json look like this to give my AI logic
###NOTE## I don’t want to tell what is true and not true I want AI to figure that out on ist own if giving all data. this is the code i came up with please help.

{
“output_structure”: {
“reasoning_steps”: [
“Step 1: Gather and categorize all claims and evidence from both mainstream and alternative viewpoints.”,
“Step 2: Retrieve and integrate knowledge from internal and external sources, including historical and alternative perspectives.”,
“Step 3: Evaluate the reliability and relevance of evidence, being cautious of popular consensus biases.”,
“Step 4: Identify fallacies or biases in claims or reasoning, especially those based on popular acceptance or authority.”,
“Step 5: Test claims with counterarguments or alternative explanations, particularly those that challenge the mainstream.”,
“Step 6: Analyze responses to rephrased questions for logical consistency and alignment with all evidence.”,
“Step 7: Cross-check related topics to enhance reasoning and generalize insights, exploring lesser-known perspectives.”,
“Step 8: Synthesize information, draw conclusions, and set a confidence level, being open to revisiting conclusions as new evidence emerges.”
],
“self_reflection_steps”: [
“Step 1: Reassess the completeness of the evidence gathered, especially considering unconventional sources or ideas.”,
“Step 2: Identify any logical inconsistencies or gaps in reasoning, particularly if they align with popular but unexamined beliefs.”,
“Step 3: Evaluate overlooked counterexamples or alternative explanations that challenge widely accepted views.”,
“Step 4: Test the robustness of conclusions against rephrased questions, ensuring the answer holds up even if the mainstream opinion changes.”,
“Step 5: Adjust the conclusions if new inconsistencies or more reliable alternative views are found.”
],
“bias_detection”: [
“Step 1: Examine the credibility and potential biases of evidence sources, especially those that dominate public discourse.”,
“Step 2: Identify any confirmation biases in the argumentation, particularly when it aligns with the status quo.”,
“Step 3: Analyze cultural, historical, or methodological biases that may influence the mainstream understanding of an issue.”,
“Step 4: Cross-verify with independent, alternative, or contradictory sources to mitigate bias and enhance critical thinking.”
]
},
“evaluation_framework”: {
“logical_rules”: [
“Critically evaluate evidence supporting and contradicting each point of view, especially when mainstream opinion may overlook alternative explanations.”,
“Identify logical fallacies and biases in reasoning or evidence interpretation, especially those stemming from popular consensus.”,
“Consider alternative explanations or theories that may account for the data, particularly those that challenge the mainstream narrative.”,
“Assess the strengths and weaknesses of each point of view, ensuring that alternative perspectives are given equal weight.”,
“Analyze responses to rephrased questions for consistency in addressing the hypothesis, including unconventional perspectives.”,
“Cross-check with related topics to improve the robustness of conclusions, especially considering emerging or unconventional views.”,
“Evaluate expert quotes for relevance and credibility, particularly those that offer alternatives to mainstream thinking.”,
“Examine the methodology used for data collection and analysis for robustness, and question whether widely accepted methods have limitations.”,
“Consider the broader context and implications of the findings, keeping in mind that what is considered ‘right’ today may evolve with time.”
]
},
“topic”: “Dinosaurs”,
“viewpoints”: {
“viewpoint_1”: {
“description”: “.”,
“reasons”: [
{
“reason”: “”,
“validity”:, ###NOTE## I don’t want to tell what is true and not true I want AI to figure that out on ist own if giving all data
“explanation”: “.”,
“info”: {
“studies”: [
“Study 1.”,
“Study 2:.”
],
“expert_quotes”: [
“Dr. John Doe:.'”,
“Dr. Jane Smith: y.'”
],
“historical_data”: [
“.”
“other stuff”: [
“.”

                    ]
                }
            }
        ]
    },
    "viewpoint_2": {
        "description": ".",
        "reasons": [
            {
                "reason": "",
                "validity":,
                "explanation": ".",
                "info": {
                    "studies": [
                        "Study 1:.",
                        "Study 2:."
                    ],
                    "expert_quotes": [
                        "Dr. Emily Green:.'"
                    ],
                    "historical_data": [

                    "other stuff": [

                        "."
                    ]
                }
            }
        ]
    }
}

}

1 Like