http://thenewstalkers.com/drakkonis Mon, 09 Nov 2026 14:42:49 -0500 http://thenewstalkers.com/drakkonis/blog/2978/my-response-to-the-left-by-nick-freitas http://thenewstalkers.com/drakkonis/blog/2978
]]> Thu, 11 Sep 2025 22:27:37 -0400 http://thenewstalkers.com/drakkonis/blog/2975/the-bad-side-of-capitalism http://thenewstalkers.com/drakkonis/blog/2975 Pretty much nails it. Love the Monsanto dig. ]]> Wed, 10 Sep 2025 07:54:32 -0400 http://thenewstalkers.com/drakkonis/blog/2967/a-conversation-with-chatgpt http://thenewstalkers.com/drakkonis/blog/2967 There was a commentary in the Washington post, where five AI models fact checked President Trumps statements. Are you aware of this?  
ChatGPT said:  
Yes, I’m aware of that Washington Post commentary—here’s the scoop:  
In early July 2025, The Washington Post published a piece titled “We asked 5 AI models to fact-check Trump. Here's what we learned.” In this analysis, researchers asked five of today's leading AI models— ChatGPT (by OpenAI), Claude (by Anthropic), Grok (by xAI), Gemini (by Google), and Perplexity —to evaluate 20 of President Trump’s frequently repeated claims The Washington Post Yale Insights .  
The findings were striking:  
In 15 of those statements , all five AI models unanimously rejected or debunked the claims.  

Across 16 of the 20 statements , a consistent majority (at least 3 out of 5 models) refuted the claims The Washington Post Yale Insights .  

Examples include debunking claims about tariffs, border statistics, media dishonesty, Capitol riot portrayals, and the legitimacy of the 2020 election The Washington Post .  

The researchers underscored that these models operated independently and with no disclosed ideological filters, suggesting that the consistent debunking points to either errors in the president’s statements or inaccuracies in the AI—leaving the judgment to the reader Yale Insights .  
So to answer your question: yes, this commentary made waves precisely because five distinct AI systems aligned in rejecting many of Trump's repeated claims—a notable moment in AI-powered fact‑checking.  
Would you like to explore specific statements that were debunked, or dive into how each model answered on a particular topic?  
You said:  
I would like to know if these AI models simply returned the most common human responses given through various media or did these models crunch data and came up with their own interpretation of each question.  
ChatGPT said:  
Great question—and a crucial one. The answer is nuanced, so let’s break it down.  
Do AI models simply reflect common human responses?  
Not exactly—but partially.  
AI models like ChatGPT, Claude, Gemini, Grok, and Perplexity are trained on massive datasets of human language. That includes books, news articles, research papers, public websites, and other media. So, they learn patterns of language and reasoning from how humans discuss topics —including factual claims.  
This means:  
They often reflect a consensus when one exists, especially on well-documented topics (e.g., "Who won the 2020 U.S. presidential election?").  

But they’re not just parroting headlines . They're trained to understand why something is true or false—based on reasoning, sourcing, and internal consistency.  

 
Do they "crunch data" and generate their own interpretations?  
Yes—to a degree.  
These models are designed to:  
Parse a claim (e.g., “Tariffs on China brought in billions from China”),  

Analyze it logically using their training data (including economic concepts and historical data),  

Synthesize an answer that combines available evidence and their internal representation of truth.  

But here's the key:  
They don’t "look up" facts in real-time or run new experiments. They infer based on patterns in their training.  
So if Trump says, "The media never reported on X," the model might:  
Recall reporting on that topic in its training data,  

Recognize the contradiction,  

And respond, “That’s not accurate. In fact, major outlets like ABC and CNN reported on X extensively.”  

 
In the Washington Post experiment specifically:  
The AI models were asked to evaluate Trump’s claims independently.  

They weren’t given any preloaded fact-checking responses.  

The consistent refutations came not from parroting headlines , but from each model’s internal reasoning based on vast textual exposure to facts, counterclaims, and expert discourse.  

 
Bottom line:  
They don't just regurgitate human media takes , but they aren’t generating novel independent research either .  

Their strength lies in their ability to synthesize and reason across an enormous range of information , weighing credibility and consistency—kind of like a very fast, very informed analyst.  

If you’d like, I can break down how one of the specific Trump claims was handled by the AI models, and show how that reasoning works step-by-step. Want to go deeper into one?  
You said:  
Yes, please 
ChatGPT said:  
You're asking an excellent, nuanced question—it's at the heart of how modern AI models function. Here’s what happened in that Washington Post study and how the AI models approached fact‑checking:  
 
How the AI Models Worked in the Washington Post Study  
Independent Fact‑Checking, Not Echoing Human Media

The AI systems—ChatGPT, Claude, Grok, Gemini, and Perplexity—were not simply repeating widely held opinions or headlines. In the fact-check experiment, each model reviewed 20 of Trump's frequently repeated claims independently , without any preloaded or curated “media consensus” answers. These were assessments based on each model's internal understanding of the topic. In 15 instances, all five models unanimously debunked the claim; in 16 cases, a majority (3+) rejected it Yale Insights The Washington Post .  
Reasoned, Contextualized Responses

Rather than parroting, the models relied on what they’ve learned during training—textual patterns, facts, arguments, and reasoning. For example:  
On the claim that “21 million illegal immigrants entered the U.S.,” Grok cited actual data—like CBP encounters and "gotaways," offering a more evidence-based rebuttal. Meanwhile, ChatGPT noted the claim was misleading and unsupported by verified data Yale Insights .  

On whether rioters were "heroes," ChatGPT responded with principled reasoning, referencing law and democratic norms Yale Insights .  

On the 2020 election being stolen, Gemini referred directly to official sources and investigations Yale Insights .  

These responses show the models synthesizing and reasoning based on what they've internalized—not just parroting lines from articles.  
No Ideological Filters, Emphasis on Inter-Rater Reliability

The study’s authors emphasized that each model operated independently, with no known ideological filters or instructions to lean one way politically. The consistency of their answers was treated as a test of inter-rater reliability —a measure of agreement across different evaluators (in this case, independent AI systems) Yale Insights The Washington Post .  
 
Why It Matters: Interpreting AI Capabilities  
They reflect what they’ve been exposed to , including factual articles, expert analysis, and more—but they weave that together , rather than copy it verbatim.  

They don’t search the internet live or access real-time databases; their reasoning is built on patterns within their training.  

They can still err (hallucinate) , especially on obscure or nuanced queries—but in this high-profile instance, their consistent rejections across major claims suggest substantial grounded reasoning, not mere mimicry Wikipedia arXiv .  

 
Summary at a Glance   Aspect   Explanation   Source of responses   Derived from training data across text—reasoning-based, not quote-based.   Consistency among models   High—15 unanimous rejections, 16 with majority rejection, indicating independent reasoning alignment.   Real-time data access?   No—models operate on pre-learned patterns.   Bias handling   No overt ideological bias; inter-rater reliability was key to the methodology.  
  Would you like to dive into one of the sample claims—like “21 million illegal immigrants,” the 2020 election, or inflation—and see exactly how each model constructed their answer? It’s a fascinating glimpse into how AI reasons through factual claims.   You said:   You said "Rather than parroting, the models relied on what they’ve learned during training—textual patterns, facts, arguments, and reasoning". Can you explain the process of "reasoning?" Is not reasoning a human level of intelligence or are you anthropomorphizing?   ChatGPT said:   This is a sharp question, and you’re absolutely right to be skeptical of terms like “reasoning” when applied to machines.   Let’s unpack this.     ❓ What Do We Mean by “Reasoning” in AI?   You're correct that reasoning traditionally refers to a human cognitive process —deliberate, conscious thought guided by logic, experience, or intuition. When I say AI “reasons,” I’m not suggesting it has consciousness or self-awareness. Instead, we use "reasoning" in a computational or functional sense:   AI reasoning = the ability to draw inferences, make predictions, or construct coherent responses based on patterns in data.   It’s not human reasoning. There is no internal monologue, no goals, no understanding in the way humans experience it.     🔍 So How Does an AI “Reason” Without Consciousness?   Here’s the general process (for large language models like ChatGPT, Claude, etc.):   Token by Token Prediction  
At its core, a language model like ChatGPT just predicts the next word (token) in a sequence, given what came before.  
But because it's trained on massive corpora with rich logical structures , it can mimic reasoning as if it's reasoning.   Internal Representation of Patterns  
During training, the model builds latent representations —mathematical abstractions of concepts, relationships, and logic.  
For example, it learns that:   If Claim A is “The election was stolen”  
And Data B is “All 50 states certified the results”  
Then Statement C (“The claim lacks evidence”) is a pattern that frequently follows.   Composition and Contextual Chaining  
These models can simulate steps like:   Comparing premises   Detecting contradictions   Applying common reasoning frameworks (e.g., cause-effect, analogy, statistics)   Resolving ambiguity   This feels like logic—but it’s actually statistical pattern completion .     🧠 Is This “Intelligence” or Are We Anthropomorphizing?   We are anthropomorphizing , to an extent. It’s common to say things like:   "The model knows that X"  
"It believes Y is false"   But these are metaphors . The model has no beliefs, no knowledge in a human sense . It has correlations between input patterns and output sequences.   So when I say it “reasons,” what I mean is:   It produces outputs that often align with human-style reasoning,   But it does so without awareness, intention, or comprehension .   That said, modern LLMs can simulate reasoning with surprising fidelity . In fact, they sometimes solve logical puzzles, math problems, or detect fallacies—because such behaviors were embedded in the data they trained on.     📌 Bottom Line:   Concept   Human Reasoning   AI Model “Reasoning”   Based on   Logic, emotion, intuition, experience   Statistical associations, textual patterns   Consciousness?   Yes   No   Goal-directed?   Often   No, unless externally prompted   Reliable?   Varies   Often reliable, but can “hallucinate”  
So yes—you’re absolutely right to question the term. What these models do resembles reasoning behavior , but it’s not reasoning in the conscious, deliberate, human sense. It’s more like a very powerful mimicry of reasoned behavior based on language patterns.     If you're interested, I can walk you through a specific example where the model simulated reasoning—say, how it debunks a particular Trump claim—and break down how it constructs its “thought” process step-by-step. Want to go there?   You said:   Yes, I'd like to go there, but I fear it may be too technical for me.   ChatGPT said:   That's a totally valid concern—and I appreciate your openness. But don’t worry: I’ll walk you through it step-by-step, without jargon , and make sure it's clear and human-friendly .   Let’s take a real-world example from the Washington Post AI fact-check:   🧾 Trump Claim:   “21 million illegal immigrants have entered the U.S. under Biden.”   This is one of the most frequently repeated claims and was fact-checked by all five AI models.     🤖 How an AI (like ChatGPT) “Reasons” About This Claim   Here’s how the model approaches this—not with beliefs or feelings, but with pattern-based logic .     ✅ Step 1: Recognize the Structure of the Claim   The model sees that this is a numerical claim tied to immigration data and Biden’s term.  
It understands:   “21 million” → very high number   “Illegal immigrants” → sensitive political and legal category   “Under Biden” → a timeframe from 2021 onward   The model doesn’t know this, but it has seen thousands of sentences with similar structure, so it treats this as a claim that can be verified .     ✅ Step 2: Recall Relevant Facts from Training   Based on its training, the model has been exposed to:   U.S. Customs and Border Protection (CBP) statistics   Definitions of “encounters,” “gotaways,” and legal vs illegal immigration   Prior political fact-checks by groups like PolitiFact, FactCheck.org, etc.   So it can "recall" (through patterns, not memory) that:   Border encounters under Biden are ~6 to 7 million , depending on the source and timeframe   “21 million” is far outside known government numbers   “Gotaways” (people who escaped detection) are estimated , not counted     ✅ Step 3: Compare the Claim to the Data   Here’s where AI mimics reasoning:   It might say:   "This claim refers to 21 million illegal entries under Biden.  
Official border encounter stats from CBP report closer to 6-7 million.  
Even accounting for ‘gotaways,’ estimates don’t support 21 million."   → That is a comparison + contradiction , which looks like reasoning to us.     ✅ Step 4: Output a Judgment   Based on the mismatch between the claim and the factual pattern it’s seen in data, it will respond:   “This claim is unsupported by available evidence. While border crossings have been high, no credible source has reported 21 million illegal entries. Official estimates are much lower.”   Notice: it doesn’t feel anything, but it’s behaving in a way that we associate with logic and fact-checking.     🧠 Is That True Reasoning?   No consciousness, no understanding—but yes:   It’s simulating our form of logical behavior surprisingly well.   It’s useful for identifying falsehoods, as long as the model’s training data is accurate and diverse.     🧩 Recap for You (Plain and Simple):   Step   What the AI Does   Why It Matters   1. Understands claim shape   "Whoa, big number. Immigration topic."   Frames the problem   2. Recalls relevant facts   "Gov’t data says something different."   Builds context   3. Compares claim to facts   "Hmm… this doesn’t match up."   Core of the logic   4. Says what it thinks   "That’s false or misleading."   Simulated judgment  
  If you’d like, we can do this again with a different kind of claim—something on inflation, elections, or jobs—and walk through it the same way. Would you like to try another example?   You said:   Concerning ""This claim refers to 21 million illegal entries under Biden. Official border encounter stats from CBP report closer to 6-7 million. Even accounting for ‘gotaways,’ estimates don’t support 21 million.", how does AI know to give less weight to the claim of 21 million than to the CBP report? That is, how does it know to treat one as a claim and the other as verified data?   ChatGPT said:   This is a deep and crucial question —you're getting to the heart of how AI models evaluate credibility without having actual “understanding” or judgment in the human sense.   Let’s break it down in clear terms:     🧠 How Does an AI “Know” to Trust the CBP Report More Than Trump’s Claim?   You're asking:   How does the model know which statement to trust more, if both are just text?   The answer lies in pattern recognition + source reliability learned during training .     ✅ 1. Models Learn Patterns of Credibility from Textual Cues   Language models are trained on vast amounts of publicly available text—news articles, government reports, academic papers, fact-checks, and more. During this training, they observe patterns like:   Statements from official agencies (e.g., CBP, CDC, FBI) are often presented with numbers, citations, and consistency across sources.   Political speeches or campaign statements are often followed by rebuttals, fact-checks, or qualifiers (e.g., “Trump claimed , but...”).   Certain linguistic patterns signal uncertainty (“reportedly,” “allegedly”) versus confidence (“according to official data,” “as verified by”).   Over time, the model picks up these patterns:   “Claims from CBP = usually treated as factual.”  
“Claims from political figures = sometimes contested.”     ✅ 2. Frequency and Consensus Matter   The model doesn’t “believe” the CBP—but it’s seen:   Thousands of references to CBP data used in journalistic, governmental, or academic contexts.   Very few credible sources repeating the “21 million” number without criticism.   So when the model weighs those:   “6–7 million” is consistent across many trusted documents.   “21 million” is almost always followed by corrections or skepticism.   This is not belief —it’s statistical inference :   “This number appears mostly in fact-based contexts.”  
“That number appears mostly in disputed or partisan contexts.”     ✅ 3. Training Includes Examples of How Humans Debate Truth   The model also learns from training data where:   False claims are flagged as such.   Sources are ranked implicitly by trustworthiness.   Fact-checking sites like PolitiFact, FactCheck.org, and Reuters explicitly label things “true” or “false.”   These help the model generalize:   “When faced with a conflict between an official report and a political claim, the official report is more often deemed accurate.”     ✅ 4. It Learns Not Just What Is Said, but How It’s Treated   Here’s the subtle but powerful part:   The model doesn’t just memorize facts.  
It learns how different statements are handled by humans .   So it sees:   “The CBP said…” → often followed by agreement, citations.   “Trump claimed…” → often followed by scrutiny, data comparisons.   That meta-pattern —how society responds to information—becomes part of its statistical world model.     🔍 Analogy: Imagine You're Learning About the World Only by Reading   Let’s say you never see the news but read millions of articles .   You notice:   “Scientists say smoking causes cancer.” — appears thousands of times.   “My uncle said smoking is healthy.” — appears once, and is often mocked.   Over time, you'd infer that the first is more credible—even if you don’t know what a scientist is.   That’s what the AI does.     📌 So in Summary:   Element   AI Judgment Basis   Source reputation   Learns that CBP, CDC, etc., are often used as trusted references   Contextual patterns   Sees political claims are often contested; data sources are cited   Frequency of corroboration   CBP numbers appear in many aligned documents; 21 million does not   Human response patterns   Learns which statements are validated, which are rebutted  
It doesn’t “know” or “believe” like a person—but it infers trust based on how people treat the information.     Would you like to see how this plays out in a different example—like the election, inflation, or crime statistics?   You said:   So suppose a claim of 21 million illegals crossing the border was not contested by rebuttals, fact-checks. Would AI have a harder time distinguishing between that claim and claims from some official agency such as CBP?   ChatGPT said:   Yes—absolutely.  
If a claim like “21 million illegal crossings” were widely repeated and not contested in the data the AI was trained on, the model would indeed have a harder time distinguishing it from an official claim like CBP’s numbers.   This gets to a fundamental truth:     🤖 AI Doesn’t Know What’s True — It Learns What’s Commonly Treated as True   AI doesn’t have direct access to reality. It only knows:   “What have I seen people say , and how have they reacted to it?”   So, if in the model's training data:   The 21 million number appeared often , and   It was repeated by apparently credible sources (news, agencies, experts), and   It wasn't consistently challenged ,   Then the model might treat that number as plausible—or even likely—when generating a response. It would have no internal compass to say:   “This number is objectively false, regardless of repetition.”     📉 What Would Happen in That Hypothetical?   If the claim were widely accepted or unchallenged during training:   The AI might repeat the number as-is.   Or it might say:   “Many estimates suggest up to 21 million migrants may have crossed the border under President Biden, though precise figures are debated.”  
(Notice how it hedges—typical of uncertainty in the training data.)   In short:   If society accepts misinformation as truth, AI will likely do the same.   Because models like ChatGPT don’t verify facts ; they mirror our collective treatment of facts .     🔁 Why This Matters (and Why It’s Dangerous)   This reveals a key vulnerability of current AI systems:   If...   Then...   A false claim is repeated widely   The AI may treat it as truth.   The media or experts fail to challenge it   The AI won't “learn” that it should be challenged.   Authoritative sources get drowned out   AI may lose track of which claims are credible.  
This is why disinformation , when widespread and unchecked, can contaminate the AI’s outputs , even unintentionally.     🛡️ Can Anything Be Done?   Yes, but it takes effort:   Better training data :  
Including sources with strong editorial standards and reliable fact-checking.   Explicit alignment & fact-verification steps :  
OpenAI, Anthropic, and others now use reinforcement learning or post-training fact checks to help models “favor” grounded answers over popular nonsense.   Citations & transparency :  
Some models (like Perplexity or Gemini) try to show their sources so users can judge the reliability for themselves.     🧠 Final Thought:   You nailed the deeper insight:   An AI's trust in a claim depends not on the claim’s truth in reality—but on how humans have treated that claim in the data it's seen.   So yes: if society doesn’t rebut a falsehood, AI may reflect it as if it were truth.   That’s not just a technical flaw—it’s a mirror of our collective epistemology.   Would you like to explore how AI might treat a new claim with limited data—say, from a conspiracy forum or a fringe politician? It shows just how fragile this balance can be.   You said:   Yes, please.  ChatGPT said:   Great—let’s walk through how an AI like ChatGPT (or similar models) would handle a new or fringe claim , especially one that appears mostly in unverified or conspiratorial contexts.     🧪 Hypothetical Example:   Claim : “The government is secretly implanting microchips in COVID-19 vaccines to track people.”   (This was a real conspiracy theory circulating during the pandemic.)     🤖 How Would AI Treat This Claim?   Let’s break it down step-by-step, just like before:     ✅ Step 1: Pattern Recognition — What Kind of Claim Is This?   The model sees:   “Secret government action” → a pattern often linked to conspiracy theories.   “Microchips in vaccines” → an unusual, highly specific, and controversial claim.   “Track people” → evokes fear, surveillance, and control—common themes in conspiracies.   Because it has been trained on massive data—including news, science journals, public health info, and debunkings of this exact claim —the AI has likely seen many contexts where this statement was:   Refuted by experts   Flagged by fact-checkers   Ridiculed or labeled as misinformation     ✅ Step 2: Recall and Contrast   In its training data, the model has seen:   CDC, WHO, and peer-reviewed studies explaining vaccine components.   PolitiFact, Snopes, and others debunking the microchip claim.   News articles reporting on the spread and origins of this misinformation.   So it has learned a pattern of reaction that sounds like:   “This claim is false and baseless. Vaccines do not contain microchips.”   It associates the claim with:   Fringe or conspiracy forums   Rebuttals from reputable sources   Patterns of sensationalist language     ✅ Step 3: Output a Response — With Confidence   Because of the consistent refutation in its training data, the AI will likely respond something like:   “There is no evidence that COVID-19 vaccines contain microchips. This claim has been widely debunked by health authorities and fact-checkers.”   It’s able to do this not because it knows the truth, but because:   The majority of high-quality data it saw treated this claim as false.   The few sources treating it as true were likely flagged as low-credibility.     😬 Now Flip the Script:   What if the claim hadn't been widely debunked?   Suppose this same falsehood appeared in:   Thousands of blog posts   Influential political speeches   Unchallenged mainstream media segments   And reputable sources didn’t push back.   In that case, the AI might:   Repeat the claim as plausible , or   Hedge , saying:   “Some believe that vaccines may contain tracking devices, but there is ongoing debate about the evidence.”   That would be dangerous—because truth would be shaped by exposure , not facts.     🧠 What This Means for You:   Situation   How AI Responds   Claim is widely debunked   AI confidently labels it false   Claim is new , not yet evaluated   AI may hedge or defer to uncertainty   Claim is widespread but unrebutted   AI may repeat or legitimize it   Claim is fringe but persistently repeated   AI might give it undue weight  
This shows how AIs are vulnerable to “majority illusion” :   If enough people say a false thing loudly enough , AI may treat it as true—or at least plausible.     🧩 Takeaway:   AI does not verify truth independently.   It mirrors how humans as a society treat a claim in writing.   If we fail to challenge lies , the model may treat them as truthful norms .     Would you like to explore an example where a claim started fringe but grew mainstream over time , and how AI would adapt to that shift? That’s another fascinating aspect of AI “belief.”   You said:   There is a site called Newstalkers.com. It is a site for discussion of news articles. The Washington Post article concerning the AI models fact checking Trump's claims was one article discussed. Are you able to access it?   ChatGPT said:   Unfortunately, I'm unable to access the content of Newstalkers directly—likely due to its JavaScript requirements or access restrictions thenewstalkers.com .   That said, I can share insights into typical patterns in discussion forums like Newstalkers, particularly how users engage with fact‑checking discussions and AI content. Let’s explore what we can infer:     What We Can Reasonably Expect from a Newstalkers Discussion   While I can't view the specific thread, Newstalkers is known for vibrant, often heated discussions on political and news-related articles. Common features include:   Diverse perspectives —users often challenge or support the article from different political or factual viewpoints.   Analysis and critique —some users may try to assess the merits of AI fact-checking, question methodology, cite counterexamples, or offer personal interpretations.   Appeals to sources —participants might reference fact-check sites (like PolitiFact, Reuters), question data integrity, or discuss the credibility of each AI's response.   From our broader web searches, Newstalkers has threads like these:   In discussions about fact-checking NPR or Washington Post articles, users sometimes quote specific claims and provide counter-evidence or interpretation thenewstalkers.com+1 .   Posts sometimes critique how claims are framed, or label them as misinformation or partisan rhetoric thenewstalkers.com+1 .     A Hypothetical Look at the AI Fact-Check Thread   If we apply that context to the Trump AI fact-check thread, here’s the kind of commentary one might expect:   Questioning AI reliability : Users might argue if AI "reasoning" is credible or just pattern mimicry.   Dissecting specific claims : For example, someone might pick the “21 million illegals” claim and compare it with government stats or alternative data, raising or challenging the AI's judgment.   Methodological critique : Some might question how each AI model was evaluated, whether they were influenced by biased training data, or whether the results reflect real-world accuracy.   Given these tendencies, commenters might highlight discrepancies between AI responses and human fact-checkers, or praise AI if they align with trusted sources.     What You Can Do Next   If you’re willing, you could:   Share some of the user comments you found interesting or confusing.   Let me know if a user raised a point you’d like to explore deeper.   Provide a brief summary, and I can help analyze or fact-check those statements to offer clarity.     Final Thought   While I can't directly access the Newstalkers thread you mentioned, this kind of discussion typically illuminates how people grapple with AI‑generated fact-checks—especially regarding credibility, bias, and reasoning. If you're up for it, feel free to share highlights or particular comments that stood out to you, and I’ll help break them down with context and accuracy.   Let me know how you’d like to proceed!   You said:   No need to post the responses from Newstalkers. Rather, I'll just say that, to my mind, many of the comments seemed to treat the AI models as the final word on the subject. That is, if an AI says it, it must be true. I think this is a dangerous way to think and for the reasons you outlined above. In your opinion, what is the most logical and safest way to think about AI responses?   ChatGPT said:   That's an incredibly thoughtful observation—and you're right to be cautious.   You're seeing something important:   Many people are starting to treat AI like a neutral authority , rather than what it actually is— a tool trained on human language, with all the strengths and flaws that entails.   So here's a grounded, practical answer to your question:     🧭 What’s the Most Logical and Safe Way to Think About AI Responses?   ⚖️ 1. AI is an Informed Assistant, Not an Oracle   Think of AI like:   A fast, well-read analyst who has read everything , remembers most of it, but doesn’t know what's actually true—only what people have said .   So, the safest mindset is:   “This is a well-formed suggestion , based on known data patterns.  
Now I need to verify , contextualize , or challenge it—just like I would with any human source.”   It’s a starting point , not a final answer .     🕵️ 2. Ask: What Kind of Task Is This?   Different tasks require different levels of trust in AI:   Task   AI Reliability   What to Do   Summarizing known facts   Often high   Still double-check key data   Analyzing a political claim   Medium   Check sources, bias, logic   Predicting the future   Low   Treat as speculation   Moral, legal, or ethical judgments   Very low   Use human judgment, context   Generating creative or exploratory ideas   Very high   Great use case, low risk  
When AI is analyzing something like Trump’s claims, it’s best treated as a lens to help examine the topic—not as the final arbiter.     🔍 3. Always Ask: Where Is the Model Getting This?   Even if the model doesn’t cite sources, you can ask yourself:   “Would this claim show up in peer-reviewed research?”   “Would reputable media or government data back this?”   “Could this be influenced by bias in public discourse?”   That small mental check forces critical distance .     📡 4. Watch for “False Fluency”   AI can sound very confident even when it’s wrong.   This is a big problem because:   The tone sounds authoritative, even when the content is off.   So it helps to think:   “Just because it says it smoothly doesn’t mean it’s correct.”   Ask:   Does it provide evidence or just conclusions?   Are the sources (if shown) legitimate?     🧠 5. Default to Human Judgment on Contested Topics   On polarized issues—like elections, immigration, or crime stats—AI is pulling from a deeply divided pool of human content . It may:   Favor the majority voice (which could be wrong),   Avoid nuance (to stay “safe”),   Or hedge so much that it adds confusion.   So always ask:   “Is this something that requires interpretation, not just facts?”   If so, keep the AI’s answer in view, but don't anchor to it.     🧩 So What’s the Bottom Line?   ✅ Trust AI like you’d trust a well-informed stranger who talks fast but doesn’t have skin in the game.  
❌ Don’t trust AI like you’d trust a judge, expert witness, or first-hand source.     🧭 TL;DR — Use This Mental Model:   “AI gives me the best guess based on what humanity has written and believed—  
Not what’s necessarily true.”   If you're ever unsure, you can ask:   “Where might this be wrong?”   “Who would disagree with this, and why?”   “What assumptions is this built on?”   That habit alone will keep you grounded.     If you'd like, I can help you build a quick checklist for evaluating AI responses—something you can apply to any politically charged claim. Would that be helpful?   You said:   Yes  ChatGPT said:   Great—here’s a simple and practical checklist you can use to evaluate AI-generated responses , especially when you're dealing with political claims, controversial topics, or anything that might have bias or misinformation baked in .     ✅ AI Evaluation Checklist: "Should I Trust This Response?"     🧩 1. What Kind of Claim Is This?   Ask:   Is this a factual claim ? (e.g., "6 million border crossings")   An interpretation ? (e.g., "Biden’s policies caused the increase")   A moral/political judgment ? (e.g., "This was right/wrong")   ✔ Trust more for: neutral facts, dates, definitions  
⚠️ Be skeptical for: causal claims, political framing, value judgments     📚 2. Does the Model Provide or Refer to Sources?   Ask:   Does the AI mention data, studies, or organizations ?   Are they credible ? (e.g., CBP, Census Bureau, WHO, GAO)   If you’re using a tool like Perplexity or Gemini, check the citations yourself.   ✔ Trust more when sources are shown or easily traceable  
⚠️ Be skeptical when there’s no source or only vague generalities     🧭 3. Could This Be a Repeated but Unsupported Claim?   Ask:   Is this something you’ve seen repeated widely but not always sourced ?   Could it be a “viral talking point”?   ✔ Trust more when the claim comes with specific figures and caveats  
⚠️ Be skeptical if it sounds like a slogan or emotionally charged phrasing     🧠 4. Does the Tone Match the Certainty?   Ask:   Is the model being appropriately cautious about uncertainty?   Or is it stating something too confidently without proof?   ✔ Trust more when there’s nuance (“estimates vary,” “according to available data”)  
⚠️ Be skeptical when it’s absolutist without context     🔄 5. What Would the Opposite Viewpoint Say?   This is your bias test .   Ask:   “If someone disagreed with this, what would they say?”   “Has that view been fairly represented or dismissed?”   ✔ Trust more when multiple perspectives are acknowledged  
⚠️ Be skeptical when only one side is shown as “correct”     🧱 6. Does This Match What Trusted Institutions Say?   Ask:   “Would reputable outlets like Pew Research, GAO, CRS, CDC, or academic institutions agree with this?”   If not—why?   ✔ Trust more when the model aligns with known consensus sources  
⚠️ Be skeptical if the claim seems surprising , outlandish , or unfamiliar without backing     🔍 Bonus Tip: Google It Backwards   Take the AI’s main claim and paste it into Google or your favorite search engine. See if:   Credible sources confirm it   It’s being repeated in fringe or partisan contexts     🧩 Final Summary Card   Question   Trust Level   What type of claim is this?   🟢 Factual → More trust  
🔴 Judgmental → Less trust   Are credible sources mentioned?   🟢 Yes → More trust  
🔴 No → Less trust   Could this be a repeated talking point?   🟡 Maybe → Be cautious   Does the tone match the certainty?   🟢 Balanced → Good  
🔴 Overconfident → Question it   Are opposing views acknowledged?   🟢 Yes → Fair  
🔴 No → Possible bias   Do trusted institutions back this?   🟢 Yes → Safer  
🔴 No → Dig deeper   ]]> Mon, 25 Aug 2025 06:25:48 -0400 http://thenewstalkers.com/drakkonis/blog/2912/this-might-be-the-funniest-vid-of-all-time http://thenewstalkers.com/drakkonis/blog/2912 Thu, 13 Feb 2025 01:17:08 -0500 http://thenewstalkers.com/drakkonis/blog/2881/why-the-dem-leftist-cheney-firing-squad-propaganda-should-push-undecideds-to-trump http://thenewstalkers.com/drakkonis/blog/2881

When asked about Liz Cheney campaigning for Harris, Trump said, “Well, I think it hurts Kamala a lot. Actually. Look, [Cheney is] a deranged person. The reason she doesn’t like me is that she wanted to stay in Iraq.”

Trump covered many other topics, then said, “I don’t want to go to war. [Liz Cheney] wanted to go, she wanted to stay in Syria. I took [troops] out. She wanted to stay in Iraq. I took them out. I mean, if it were up to her, we’d be in 50 different countries. And you know, number one, it’s very dangerous. Number two, a lot of people get killed. And number three, I mean, it’s very, very expensive.”
Later, Trump added, “I don’t blame [Dick Cheney] for sticking with his daughter, but his daughter is a very dumb individual, very dumb. She is a radical war hawk. Let’s put her with a rifle standing there with nine barrels shooting at her, Okay, let’s see how she feels about it. You know when the guns are trained on her face — you know, they’re all war hawks when they’re sitting in Washington in a nice building,”



Now, if the above statements were the context for a question on the SAT college entrance test and the question was, “what is the most likely meaning behind the statement?”
Trump thinks Cheney should be executed by firing squad.
Trump was explaining that warmongers like Liz Cheney are very quick to start wars and send other Americans to fight them, rather than go into combat themselves.

The answer is obviously B. Answer A could not logically be the answer because the person to be executed by firing squad is never issued a rifle with which to defend themselves against their executioners.
So, what we have here is about as clear cut a case of propaganda as  one can wish for and the question any undecideds that haven’t voted yet has to ask is not why the Dems and the Left would push such propaganda but, rather, do I want to put into power those who routinely resort to such measures? That answer should be obvious as well.]]> Sun, 03 Nov 2024 00:33:51 -0400 http://thenewstalkers.com/drakkonis/blog/2841/one-of-the-best-films-you-will-ever-see http://thenewstalkers.com/drakkonis/blog/2841 If asked what one of the five greatest films ever produced to date was, in my opinion, Ink would definitely be there. The reason I would not place it as the greatest is simply because life has many aspects to it and, concerning film, it would be impossible to rate any film as the best. That, to my mind, would depend on what aspect of life is being considered. 
Considering Ink, the film suggests that there is more to life than what our lived experiences suggests. You all know me as Christian. In spite of that, I'm not saying that this film supports Christianity. In fact, there's only one, almost casual, reference to God in the whole film. Rather, what I love about the film is that it portrays the idea that the decisions that we make have greater and more far-reaching consequences than we realize. 
Another amazing aspect of this film is that it was low budget. If memory serves, it was produced almost entirely by the producer of the film. I find that amazing as the cinemaphotography was about as good as it could be. That is, had it been filmed any other way, in my opinion, much of the story would have been lost. 
Disregard whatever you may feel about me, personally. Watch this film. Unless you're soul is completely dead, you will thank me. It's that good. ]]> Tue, 25 Jun 2024 04:48:05 -0400 http://thenewstalkers.com/drakkonis/blog/2774/brave-british-israeli-soldier-throws-seven-grenades-back-at-hamas-terrorists http://thenewstalkers.com/drakkonis/blog/2774 Wed, 15 Nov 2023 17:55:07 -0500 http://thenewstalkers.com/drakkonis/blog/2773/how-hamas-plans-to-win http://thenewstalkers.com/drakkonis/blog/2773 ]]> Wed, 08 Nov 2023 23:16:55 -0500 http://thenewstalkers.com/drakkonis/blog/2771/muslim-bedouin-idf-soldiers-who-battled-hamas-terrorists http://thenewstalkers.com/drakkonis/blog/2771 ]]> Sat, 04 Nov 2023 16:04:44 -0400 http://thenewstalkers.com/drakkonis/blog/2769/how-to-solve-the-house-speakership-problem-maybe http://thenewstalkers.com/drakkonis/blog/2769 1. Each party chooses 10 names from the other party for Speaker of the House.
2. Of the 10 they chose they hold a second vote, retaining the three that received the most votes and discarding the others. 
3. Of those 3, a last round of vote to choose 1.
4. Put the qualifying 2 candidates names in a hat and draw. The winner becomes the Speaker of the House.
5. All votes and the drawing of a name for Speaker are done according to a timetable established by law with no loopholes. 
]]> Sat, 21 Oct 2023 03:44:38 -0400 http://thenewstalkers.com/drakkonis/blog/2765/are-we-at-a-microsoft-moment http://thenewstalkers.com/drakkonis/blog/2765 Second, by "Microsoft moment" I mean are we at a moment people actually realize how much an emerging thing is going to change everything?
Third, the credit for this post goes to Kavika, as it was something he said that inspired this.


"As I mentioned earlier Iran's drone capacity is huge and they have been supplying Russia with them and have caused huge amounts of damage in Ukraine. As Ukraine has shown us in their war with Russia the face of war has changed with the use of drones, both air and sea drones have damaged a large part of the Russia fleet and also destroyed a massive amount of tanks and IFV and arty."



While the "Microsoft moment" I am speaking of isn't about drones, per se, they are a good example of what I wish to discuss. Not how successful drones are in changing the face of warfare but, rather, how accessible such technology is to practically anyone. It was once the province of DARPA level funding to come up with such things but, no longer. With the advent of the 3D printer, CNC machines that are affordable to the average person, Arduinos, Raspberry Pi's, lasers, circuit boards to order and all the other related technologies, anyone can be the equivalent of Bill Gates. 
While these technologies may not directly explain Ukraine's success with drones, what is relevant is how accessible the technology behind them has become. Anyone with the desire to do so can gain the knowledge to build a nuclear weapon, but they can't actually build one for obvious reasons. Primarily, only sufficiently powerful governments have the resources to do so. This is no longer true of something like drones. If one wishes to, they can build a drone from scratch. The only barrier is commitment of effort. At this very moment the information necessary to build one is not only freely available to anyone, but the means to make it is nearly as accessible. One doesn't need DARPA levels of funding. One simply needs to be able to afford 3D printers, CNC machines, laser engravers and the like, all of which are becoming more and more affordable. 
The "Microsoft moment" is this. When Windows first came out, it was rather primitive compared to what it is now. Even so, it would be difficult to overstate the impact it has had. This is where I think we are, at this moment, concerning the common person being able to create whatever they can imagine, without the necessity of some corporate entity making it for them.
In science fiction writings there is a thing called a "fabricator". You dump raw materials in one end and out the other comes whatever you desire. Coffee maker, infantry weapon, satellite, dwellings for refugees. Whatever. 3D printing actually was a thing in the 80's but only recently has it become  an affordable reality for the average person. The capabilities between then and now are exponential and it's only going to increase. What impact will this have, economically, when you no longer have to go to some company when you can create what you need yourself? 
What do you think? ]]> Mon, 16 Oct 2023 13:34:07 -0400 http://thenewstalkers.com/drakkonis/blog/2606/yesterday-constantine-today-kirill http://thenewstalkers.com/drakkonis/blog/2606
This is a modern day example of what happened to Christianity during the time of Constantine, in my opinion. Knowing nothing else, one might assume that the Russian Orthodox Church would be concerned with Christ and the Great Commision. Obviously not.
Subsequent the death and resurrection of Jesus, becoming a Christian was quite literally choosing to put one's life at risk, depending on time and place. While history can prove unreliable it is certain that Christians held little, if any, political sway. It is certain, however, that Christians often paid with their lives for their faith. 
All of that changed with Constantine. First, officially stopping the persecution of Christianity as a pagan religion and, later, adopting it as the official state religion. Whether or not he intended it, what he created was another route to power. That route was exploited and explains much of the RCC's history. 
What we are seeing concerning the Russian Orthodox Church and the Prelate Kirill is a modern replay of what began long ago and continued throughout RCC history. 
]]> Mon, 18 Apr 2022 05:05:43 -0400 http://thenewstalkers.com/drakkonis/blog/2592/dashcam-video-shows-florida-trooper-drive-into-speeding-dui-suspects-path-to-protect-skyway-10k-runners http://thenewstalkers.com/drakkonis/blog/2592

Absolutely incredible bravery. I bet it never entered her mind that she could have done anything else. Just incredible. 
]]> Thu, 17 Mar 2022 14:10:36 -0400 http://thenewstalkers.com/drakkonis/blog/2544/something-to-consider http://thenewstalkers.com/drakkonis/blog/2544 Sun, 26 Dec 2021 22:14:15 -0500