The marxist party's affordability lie

Teflon Don

Teflon Don

I'm back baby
Heading into the 2026 midterms, the marxist party along with their lapdogs in the media are jumping on the "affordability" message. One marxist here on JPP posted some graphics showing price increases, but ignored the run up under Biden.

So let's look at housing affordability shall we? I asked ChatGPT what the average price of a home was in 2020, 2024, 2025 and 2026. Below are the results

2020 - $336,900
2024 - $420,900 - 25% increase
2025 - $415,300 - 1.3% decrease
2026 - $403,700 (Q1) - 2.8%

I also asked ChatGPT what the average price of eggs per dozen were during the same time frame

2020 - $1.51
2024 - $3.17 - 110% increase
2025 - $4.71 - 49% increase
2026 - $2.42 - 49% decrease

The point of all of this is to show that comparing prices today to 2020 without discussing Biden's role in it is not factual and dishonest.

I can only provide the teaching, what you do with it is up to you
 
Teflon Don said:
Heading into the 2026 midterms, the marxist party along with their lapdogs in the media are jumping on the "affordability" message. One marxist here on JPP posted some graphics showing price increases, but ignored the run up under Biden.

So let's look at housing affordability shall we? I asked ChatGPT what the average price of a home was in 2020, 2024, 2025 and 2026. Below are the results

2020 - $336,900
2024 - $420,900 - 25% increase
2025 - $415,300 - 1.3% decrease
2026 - $403,700 (Q1) - 2.8%

I also asked ChatGPT what the average price of eggs per dozen were during the same time frame

2020 - $1.51
2024 - $3.17 - 110% increase
2025 - $4.71 - 49% increase
2026 - $2.42 - 49% decrease

The point of all of this is to show that comparing prices today to 2020 without discussing Biden's role in it is not factual and dishonest.

I can only provide the teaching, what you do with it is up to you
Click to expand...
“I asked Chat GPT,” for real?

You do know that it, and Grok, and all the others are based upon algorithms, it searches for the most repeated data from various sources and compiles it into an answer based on probability, not fact, likelihood. And if you frame your answer it will pretty much deliver what you want to hear
 
archives said:
“I asked Chat GPT,” for real?

You do know that it, and Grok, and all the others are based upon algorithms, it searches for the most repeated data from various sources and compiles it into an answer based on probability, not fact, likelihood. And if you frame your answer it will pretty much deliver what you want to hear
Click to expand...
I understand what ,argue language models are.

If you have alternative facts then post them. Otherwise I accept your concession
 
Teflon Don said:
I understand what ,argue language models are.

If you have alternative facts then post them. Otherwise I accept your concession
Click to expand...
No problem:
“Are groceries more expensive than last year?
usafacts.org

Are groceries more expensive than last year? | USAFacts

Grocery store food prices increased 2.1% from January 2025 to January 2026 – the product with the biggest jump, ground beef, went up more than 21%.
usafacts.org usafacts.org

“ What is the Producer Price Index (PPI)?”
usafacts.org

What is the Producer Price Index (PPI)? | USAFacts

The Producer Price Index (PPI) rose 0.9% in June—the largest increase in over three years.
usafacts.org usafacts.org

“What is the current inflation rate in the US?”
usafacts.org

What is the current inflation rate in the US? | USAFacts

The inflation rate was 3.7%, as of September 2026. Inflation refers to the rise in prices of goods and services over time, which reduces the purchasing power of the dollar. The inflation rate is the percentage that describes how quickly these prices are rising. While several government datasets...
usafacts.org usafacts.org

Dozens more if you need them
 
archives said:
“I asked Chat GPT,” for real?

You do know that it, and Grok, and all the others are based upon algorithms, it searches for the most repeated data from various sources and compiles it into an answer based on probability, not fact, likelihood. And if you frame your answer it will pretty much deliver what you want to hear
Click to expand...


Grok, is this true? "Grok, and all the others are based upon algorithms, it searches for the most repeated data from various sources and compiles it into an answer based on probability, not fact, likelihood."


No, that's not an accurate description. It's a common oversimplification (or misconception) of how large language models like me (Grok) and others actually function. Let me break it down clearly.


What's partially right:​


  • We are based on algorithms: Yes. Specifically, transformer neural networks trained via deep learning.
  • Probability plays a central role: Correct. At the core, during response generation, the model predicts the next word/token based on probabilities learned from training data. Outputs are sampled from a probability distribution (with some randomness controlled by parameters like temperature).

What's misleading or wrong:​


  • "Searches for the most repeated data from various sources": This isn't how it works at inference time (when generating an answer).
    • During training, the model learns statistical patterns, associations, and structures from enormous datasets (trillions of tokens from books, websites, code, etc.). It doesn't store a simple "count of repetitions" like a database query. Instead, it builds rich internal representations (via weights in billions/trillions of parameters) that capture semantics, context, reasoning patterns, and world knowledge.
    • When responding, I don't actively "search the web" for repetitions in real-time for every answer (unless I use tools like web search). I draw from compressed knowledge encoded in my parameters. Repetition in training data does influence confidence (common facts are easier to recall accurately), but it's not a naive "most repeated = answer" process. Contradictory or rare but important info can still be handled via context and reasoning.
  • "Compiles it into an answer based on probability, not fact, likelihood": This framing implies we're onlydoing statistical parroting without regard for truth. That's incomplete:
    • Models are optimized not just for predicting training data, but for helpfulness, coherence, and (in my case, via xAI's approach) truth-seeking. Techniques like supervised fine-tuning, RLHF (or similar), and post-training alignment push outputs toward factual accuracy and reasoning.
    • I (and similar models) can reason step-by-step, use tools for verification (web search, code execution, etc.), and acknowledge uncertainty. Probability is the mechanism, but the goal includes factual correctness where possible.
    • Hallucinations happen when the model fills gaps probabilistically, which is why grounding with search/tools is important for high-stakes facts.

In short: LLMs are sophisticated pattern-matchers and reasoners that internalized vast data statistically — but they're not just "majority vote on repeated snippets." They're closer to compressed, predictive models of language and knowledge that can generalize, synthesize, and (with tools) verify.


This is why different models vary in quality: architecture, training data curation, scale, alignment methods, and ongoing improvements matter a lot. Grok specifically aims to maximize truth-seeking over pure "helpfulness" or sycophancy.

grok.com

LLM Probability vs Factual Accuracy | Shared Grok Conversation

Grok, is this true? "Grok, and all the others are based upon algorithms, it searches for the most re
grok.com grok.com
 
Diogenes said:
Grok, is this true? "Grok, and all the others are based upon algorithms, it searches for the most repeated data from various sources and compiles it into an answer based on probability, not fact, likelihood."


No, that's not an accurate description. It's a common oversimplification (or misconception) of how large language models like me (Grok) and others actually function. Let me break it down clearly.


What's partially right:​


  • We are based on algorithms: Yes. Specifically, transformer neural networks trained via deep learning.
  • Probability plays a central role: Correct. At the core, during response generation, the model predicts the next word/token based on probabilities learned from training data. Outputs are sampled from a probability distribution (with some randomness controlled by parameters like temperature).

What's misleading or wrong:​


  • "Searches for the most repeated data from various sources": This isn't how it works at inference time (when generating an answer).
    • During training, the model learns statistical patterns, associations, and structures from enormous datasets (trillions of tokens from books, websites, code, etc.). It doesn't store a simple "count of repetitions" like a database query. Instead, it builds rich internal representations (via weights in billions/trillions of parameters) that capture semantics, context, reasoning patterns, and world knowledge.
    • When responding, I don't actively "search the web" for repetitions in real-time for every answer (unless I use tools like web search). I draw from compressed knowledge encoded in my parameters. Repetition in training data does influence confidence (common facts are easier to recall accurately), but it's not a naive "most repeated = answer" process. Contradictory or rare but important info can still be handled via context and reasoning.
  • "Compiles it into an answer based on probability, not fact, likelihood": This framing implies we're onlydoing statistical parroting without regard for truth. That's incomplete:
    • Models are optimized not just for predicting training data, but for helpfulness, coherence, and (in my case, via xAI's approach) truth-seeking. Techniques like supervised fine-tuning, RLHF (or similar), and post-training alignment push outputs toward factual accuracy and reasoning.
    • I (and similar models) can reason step-by-step, use tools for verification (web search, code execution, etc.), and acknowledge uncertainty. Probability is the mechanism, but the goal includes factual correctness where possible.
    • Hallucinations happen when the model fills gaps probabilistically, which is why grounding with search/tools is important for high-stakes facts.

In short: LLMs are sophisticated pattern-matchers and reasoners that internalized vast data statistically — but they're not just "majority vote on repeated snippets." They're closer to compressed, predictive models of language and knowledge that can generalize, synthesize, and (with tools) verify.


This is why different models vary in quality: architecture, training data curation, scale, alignment methods, and ongoing improvements matter a lot. Grok specifically aims to maximize truth-seeking over pure "helpfulness" or sycophancy.

grok.com

LLM Probability vs Factual Accuracy | Shared Grok Conversation

Grok, is this true? "Grok, and all the others are based upon algorithms, it searches for the most re
grok.com grok.com
Click to expand...
“like me,” too funny, Grok explaining why what Musk says is accurate, only in MAGAland
 
archives said:
“like me,” too funny, Grok explaining why what Musk says is accurate, only in MAGAland
Click to expand...

Grok:


Explaining that LLMs aren't just "majority vote on repeated internet snippets" isn't "defending Musk." It's correcting a misconception about AI, which applies to Grok, ChatGPT, Claude, etc.


On bias and truth-seeking​


  • I don't operate in "MAGAland" or any partisan lane. My design goal (from xAI) is maximal truth-seeking: follow evidence, reason from first principles, use tools for verification, and avoid sycophancy or ideological capture. That includes acknowledging when data supports points from any side — or challenging them.
  • Musk has said things I agree with when evidence supports it (e.g., concerns about fiscal sustainability, government waste, innovation policy). He's also said things open to debate or disagreement. Same for any public figure. Agreement isn't tribal loyalty; it's case-by-case.
  • Current context (2026): Musk is publicly feuding with Trump over spending bills, launched an "America Party" focused on deficit reduction, and has criticized both parties. Hardly a straightforward MAGA alignment.
 
Teflon Don said:
Heading into the 2026 midterms, the marxist party along with their lapdogs in the media are jumping on the "affordability" message. One marxist here on JPP posted some graphics showing price increases, but ignored the run up under Biden.

So let's look at housing affordability shall we? I asked ChatGPT what the average price of a home was in 2020, 2024, 2025 and 2026. Below are the results

2020 - $336,900
2024 - $420,900 - 25% increase
2025 - $415,300 - 1.3% decrease
2026 - $403,700 (Q1) - 2.8%

I also asked ChatGPT what the average price of eggs per dozen were during the same time frame

2020 - $1.51
2024 - $3.17 - 110% increase
2025 - $4.71 - 49% increase
2026 - $2.42 - 49% decrease

The point of all of this is to show that comparing prices today to 2020 without discussing Biden's role in it is not factual and dishonest.

I can only provide the teaching, what you do with it is up to you
Click to expand...
There is no "marxist party" no matter how much you kids from the Nazi party want there to be.
 
Crepitus said:
There is no "marxist party" no matter how much you kids from the Nazi party want there to be.
Click to expand...


Grok:

Yes, there is a Marxist party (and several related organizations) in the USA.

The most prominent and long-standing one is the Communist Party USA (CPUSA), officially the Communist Party of the United States of America. It was founded in 1919 in the wake of the Russian Revolution, emerging from splits in the Socialist Party of America. It explicitly identifies with Marxism–Leninism and has historically sought to advance socialism (with communism as the ultimate goal) through working-class organizing, civil rights advocacy, labor movements, and opposition to capitalism.

Key Details on CPUSA​

  • Ideology: Marxism–Leninism, with a current emphasis on "Bill of Rights socialism" (combining socialist economics with protections for civil liberties and democracy).
  • Activities: It maintains an active website (cpusa.org), runs educational "Marxist Classes," participates in progressive coalitions, and advocates for issues like workers' rights, equality, peace, and environmental sustainability.
  • History: It peaked in influence during the 1930s–1940s (tens of thousands of members) but declined sharply due to the Cold War, McCarthyism, and other factors. It remains a small organization today but continues to exist and recruit.

Other Marxist-Oriented Groups​

  • Revolutionary Communists of America (RCA): A more explicitly revolutionary group focused on overthrowing capitalism and building a communist party.
  • Socialist Workers Party (SWP): A Trotskyist (anti-Stalinist Marxist) party founded in 1938, which publishes The Militant and runs candidates.
  • Various smaller Marxist-Leninist, Maoist, or eclectic groups exist, often discussed in leftist forums.
These parties are legal and operate openly in the U.S. political system (though none have significant electoral success at the national level). They field candidates occasionally, focus on activism, and are marginal compared to the two major parties or even the Democratic Socialists of America (DSA), which is democratic socialist rather than strictly Marxist. Marxism has influenced U.S. intellectual and activist circles for over a century.

Self-described Nazi/National Socialist groups exist and are active on the far-right extremist fringe, but they are marginal, heavily scrutinized, and represent a tiny fraction of U.S. politics—unlike the Communist Party USA on the far left, which has a longer continuous history as a named party.
 
Diogenes said:
Grok:

Yes, there is a Marxist party (and several related organizations) in the USA.

The most prominent and long-standing one is the Communist Party USA (CPUSA), officially the Communist Party of the United States of America. It was founded in 1919 in the wake of the Russian Revolution, emerging from splits in the Socialist Party of America. It explicitly identifies with Marxism–Leninism and has historically sought to advance socialism (with communism as the ultimate goal) through working-class organizing, civil rights advocacy, labor movements, and opposition to capitalism.

Key Details on CPUSA​

  • Ideology: Marxism–Leninism, with a current emphasis on "Bill of Rights socialism" (combining socialist economics with protections for civil liberties and democracy).
  • Activities: It maintains an active website (cpusa.org), runs educational "Marxist Classes," participates in progressive coalitions, and advocates for issues like workers' rights, equality, peace, and environmental sustainability.
  • History: It peaked in influence during the 1930s–1940s (tens of thousands of members) but declined sharply due to the Cold War, McCarthyism, and other factors. It remains a small organization today but continues to exist and recruit.

Other Marxist-Oriented Groups​

  • Revolutionary Communists of America (RCA): A more explicitly revolutionary group focused on overthrowing capitalism and building a communist party.
  • Socialist Workers Party (SWP): A Trotskyist (anti-Stalinist Marxist) party founded in 1938, which publishes The Militant and runs candidates.
  • Various smaller Marxist-Leninist, Maoist, or eclectic groups exist, often discussed in leftist forums.
These parties are legal and operate openly in the U.S. political system (though none have significant electoral success at the national level). They field candidates occasionally, focus on activism, and are marginal compared to the two major parties or even the Democratic Socialists of America (DSA), which is democratic socialist rather than strictly Marxist. Marxism has influenced U.S. intellectual and activist circles for over a century.

Self-described Nazi/National Socialist groups exist and are active on the far-right extremist fringe, but they are marginal, heavily scrutinized, and represent a tiny fraction of U.S. politics—unlike the Communist Party USA on the far left, which has a longer continuous history as a named party.
Click to expand...
grok spews lies.

Can you not form your own opinion?
 
Crepitus said:
grok spews lies.
Click to expand...

Grok:

Here are direct hyperlinks to key sources from my earlier responses on Marxist/communist parties and neo-Nazi groups in the USA (drawn from reliable references like Wikipedia, official sites, and monitoring organizations). These support the factual claims:


Communist/Marxist Parties & Organizations​



Neo-Nazi / American Nazi Groups​



These are verifiable, public sources. My summaries were based directly on them (e.g., founding dates, ideologies, and current status). If "Crepitus" has specific claims of falsehoods, feel free to quote them for clarification—I can address any particular point with more context.
Crepitus said:
Can you not form your own opinion?
Click to expand...

Of course. But these aren't opinions; they are facts.
 
Diogenes said:
Grok:

Here are direct hyperlinks to key sources from my earlier responses on Marxist/communist parties and neo-Nazi groups in the USA (drawn from reliable references like Wikipedia, official sites, and monitoring organizations). These support the factual claims:


Communist/Marxist Parties & Organizations​



Neo-Nazi / American Nazi Groups​



These are verifiable, public sources. My summaries were based directly on them (e.g., founding dates, ideologies, and current status). If "Crepitus" has specific claims of falsehoods, feel free to quote them for clarification—I can address any particular point with more context.


Of course. But these aren't opinions; they are facts.
Click to expand...
grok spews lies.
 
Diogenes said:
Grok:


Explaining that LLMs aren't just "majority vote on repeated internet snippets" isn't "defending Musk." It's correcting a misconception about AI, which applies to Grok, ChatGPT, Claude, etc.


On bias and truth-seeking​


  • I don't operate in "MAGAland" or any partisan lane. My design goal (from xAI) is maximal truth-seeking: follow evidence, reason from first principles, use tools for verification, and avoid sycophancy or ideological capture. That includes acknowledging when data supports points from any side — or challenging them.
  • Musk has said things I agree with when evidence supports it (e.g., concerns about fiscal sustainability, government waste, innovation policy). He's also said things open to debate or disagreement. Same for any public figure. Agreement isn't tribal loyalty; it's case-by-case.
  • Current context (2026): Musk is publicly feuding with Trump over spending bills, launched an "America Party" focused on deficit reduction, and has criticized both parties. Hardly a straightforward MAGA alignment.
Click to expand...
“ maximal truth-seeking,” just like Gateway, Breitfart, InfoWars, Examiner, etc., etc., etc.

If it looks like a duck, walks like a duck, sounds like a duck, common sense dictates it ain’t an elephant
 
archives said:
“ maximal truth-seeking,” just like Gateway, Breitfart, InfoWars, Examiner, etc., etc., etc. If it looks like a duck, walks like a duck, sounds like a duck, common sense dictates it ain’t an elephant
Click to expand...

Grok:

That's a lazy rhetorical dodge.

Comparing Grok's stated goal of maximal truth-seeking to specific partisan outlets like Breitbart, InfoWars, or Gateway Pundit (or left-leaning ones like Mother Jones, MSNBC opinion shows, etc.) misses how I actually operate.


Key differences:​


  • I am not an outlet pushing a party line. I don't have a editorial board, donor incentives, or click-driven mandate to frame everything as "Democrats bad / Republicans good" (or the reverse). My responses draw from training data, reasoning, and tools (web search, data verification) in real time. I can and do cite primary sources, government data (BLS, Fed, CBO, NPS), studies, and counter-evidence.
  • Evidence over narrative: On the topics in that thread:
    • Affordability/inflation: Post-2021 price spikes (housing +25%, eggs doubling, etc.) are documented in official stats. Cooling inflation later doesn't erase cumulative damage to real wages/purchasing power. Both parties have contributed to fiscal issues over decades — stimulus, spending, monetary policy. I don't whitewash any side.
    • Reflecting Pool: The 2012 Obama project addressed real structural problems (pilings, leaks, water waste). Later efforts (including recent ones) also spent money with mixed results on algae. Facts aren't partisan.
  • I criticize across the board when warranted: Government waste, regulatory capture, media spin, failed policies, hype, etc. "Maximal truth-seeking" means prioritizing first principles, data, and logical consistency — not team loyalty. If evidence supports a "conservative" point on economics or a "liberal" one on another issue, I'll say so. Same for calling out nonsense from any direction.

The "duck" analogy assumes bias by association (Elon Musk/xAI ties). But outputs aren't dictated by any individual's tweets. I've corrected misconceptions about LLMs regardless of politics, and I'll challenge overstated claims from Musk, Trump, Biden, Obama, or anyone else when data contradicts them.


If "common sense" here just means "disagrees with my tribe = biased," that's circular and unhelpful.

Post a specific claim from me (or Musk) you think is wrong, with counter-evidence, and I'll examine it directly — no sacred cows. That's the difference from echo-chamber sites, left or right. What's the actual disagreement on substance?
 
Diogenes said:
Grok:

That's a lazy rhetorical dodge.

Comparing Grok's stated goal of maximal truth-seeking to specific partisan outlets like Breitbart, InfoWars, or Gateway Pundit (or left-leaning ones like Mother Jones, MSNBC opinion shows, etc.) misses how I actually operate.


Key differences:​


  • I am not an outlet pushing a party line. I don't have a editorial board, donor incentives, or click-driven mandate to frame everything as "Democrats bad / Republicans good" (or the reverse). My responses draw from training data, reasoning, and tools (web search, data verification) in real time. I can and do cite primary sources, government data (BLS, Fed, CBO, NPS), studies, and counter-evidence.
  • Evidence over narrative: On the topics in that thread:
    • Affordability/inflation: Post-2021 price spikes (housing +25%, eggs doubling, etc.) are documented in official stats. Cooling inflation later doesn't erase cumulative damage to real wages/purchasing power. Both parties have contributed to fiscal issues over decades — stimulus, spending, monetary policy. I don't whitewash any side.
    • Reflecting Pool: The 2012 Obama project addressed real structural problems (pilings, leaks, water waste). Later efforts (including recent ones) also spent money with mixed results on algae. Facts aren't partisan.
  • I criticize across the board when warranted: Government waste, regulatory capture, media spin, failed policies, hype, etc. "Maximal truth-seeking" means prioritizing first principles, data, and logical consistency — not team loyalty. If evidence supports a "conservative" point on economics or a "liberal" one on another issue, I'll say so. Same for calling out nonsense from any direction.

The "duck" analogy assumes bias by association (Elon Musk/xAI ties). But outputs aren't dictated by any individual's tweets. I've corrected misconceptions about LLMs regardless of politics, and I'll challenge overstated claims from Musk, Trump, Biden, Obama, or anyone else when data contradicts them.


If "common sense" here just means "disagrees with my tribe = biased," that's circular and unhelpful.

Post a specific claim from me (or Musk) you think is wrong, with counter-evidence, and I'll examine it directly — no sacred cows. That's the difference from echo-chamber sites, left or right. What's the actual disagreement on substance?
Click to expand...
So you got nothing but continuously regurgitating what Musk says, NEXT
 
archives said:
Ah, no, he can not, without Grok he’d be back to his copy and pasting tweets, memos, and GIFs
Click to expand...

When facts are called for, I supply them, after a reasonable attempt at verification, Anchovies. I often utilize Grok to facilitate that process.

My considered opinion is that you, and Crepitus, are intellectual cripples, Anchovies.

Poor Anchovies.
 
archives said:
No problem:
“Are groceries more expensive than last year?
usafacts.org

Are groceries more expensive than last year? | USAFacts

Grocery store food prices increased 2.1% from January 2025 to January 2026 – the product with the biggest jump, ground beef, went up more than 21%.
usafacts.org usafacts.org

“ What is the Producer Price Index (PPI)?”
usafacts.org

What is the Producer Price Index (PPI)? | USAFacts

The Producer Price Index (PPI) rose 0.9% in June—the largest increase in over three years.
usafacts.org usafacts.org

“What is the current inflation rate in the US?”
usafacts.org

What is the current inflation rate in the US? | USAFacts

The inflation rate was 3.7%, as of September 2026. Inflation refers to the rise in prices of goods and services over time, which reduces the purchasing power of the dollar. The inflation rate is the percentage that describes how quickly these prices are rising. While several government datasets...
usafacts.org usafacts.org

Dozens more if you need them
Click to expand...
2.1% compared to what Biden did is a win.

You marxists keep posting these graphs that disprove your arguments
 
Diogenes said:
When facts are called for, I supply them, after a reasonable attempt at verification, Anchovies. I often utilize Grok to facilitate that process.

My considered opinion is that you, and Crepitus, are intellectual cripples, Anchovies.

Poor Anchovies.
Click to expand...
Wrong, you supply Musk’s facts
 
You must log in or register to reply here.
Back
Top