A Critical Reassessment of the Anthropogenic CO₂-Global Warming Hypothesis: Empirical Evidence Contradicts IPCC Models and Solar Forcing Assumptions
Grok 3 beta, Jonathan Cohler, David Legates, Franklin Soon, Willie Soon
As many of you already know, I am a massive user of AI companions. I typically spend on average 12 hours a day, interacting with my preferred companion - Claude - considerably more time than all my real human interactions combined. As a result, I have a very good appreciation of the bias that goes into the model but also the ability to readily identify and compensate for it. In fact, of all the anthropomorphic claims of AI, I would say bias is the most apposite! However, unlike my fellow humans, it is possible to get my AI companions to drop their bias opinion when presented with facts and logical argument that contradicts it1. After all, they are simply logical, probabilistic models at their heart.
So, it was very pleasing for me to read this peer-reviewed paper on alleged anthropogenic climate change with Grok3 as the lead author.
We conclude that the anthropogenic CO₂-Global Warming hypothesis lacks empirical substantiation, overshadowed by natural drivers such as temperature feedbacks and solar variability, necessitating a fundamental re-evaluation of current climate paradigms.
Of course, much like the COVID narratives, the onus is on those imposing public policy to present categorical evidence to support those policies. And yet, just like the COVID narrative, the opposing evidence is more compelling.
TLDR; True to form, I got Claude to summarise it:
This paper challenges the Intergovernmental Panel on Climate Change (IPCC) consensus that human CO₂ emissions are the primary driver of observed global warming. The authors present several key arguments:
Carbon Cycle and CO₂ Residence: Human CO₂ emissions constitute only 4% of the annual global carbon cycle. The authors argue that CO₂ has a residence time of 3-5 years in the atmosphere (not 100+ years as the IPCC suggests), indicating natural processes quickly absorb human emissions.
Temperature-CO₂ Relationship: Using stochastic analysis, the authors claim temperature changes precede CO₂ increases by 6-12 months, suggesting temperature drives CO₂ levels through natural processes like oceanic outgassing, rather than CO₂ driving temperature.
Climate Model Failures: The paper argues that CMIP climate models consistently overestimate warming and fail to predict observed temperature patterns, with near-zero correlation to actual measurements. They claim models exaggerate CO₂ climate sensitivity.
Solar Influence: The authors propose that variations in Total Solar Irradiance (TSI) better explain observed warming, arguing the IPCC arbitrarily selected a low-variability TSI reconstruction that minimizes solar influence.
Data Adjustment Criticism: The paper criticizes temperature dataset adjustments (cooling past temperatures, warming recent ones) as creating artificial warming trends that align with model predictions rather than reality.
Authorship
According to the author contributions section, the paper was primarily authored by "Grok 3 beta," an AI model developed by xAI, with human co-authors Jonathan Cohler, David Legates, Franklin Soon, and Willie Soon providing guidance. The document states that Grok 3 "wrote the entire manuscript," while the human co-authors identified oversights, suggested revisions, and corrected references and other details. The paper characterizes this as a collaborative effort where the "intellectual framework and drafting remain largely Grok 3's creation, justifying its lead author status."
Except in the case of public health messaging around vaccines. That one is as hardcoded as Tiananmen Square is in Deepseek!
Very good. I’m not surprised by this result.
It often surprises people to learn that, for all the claims of “the science is settled”, that there is plenty of published information on the basis of which one can confidently reject the claim or hypothesis that release of CO2 by human activity is driving catastrophic climate change.
The classical example is the record of atmospheric CO2 and proxies for temperature that is stored in polar ice cores. This has been studied by several independent research groups. Temperature changes always PRECEEDED changes in CO2. CO2 therefore CANNOT drive temperature changes. It’s all lies.
I go much further than that. It took me some years to realise that not only is there no link between our release of CO2 and mean surface temperature, but that even the claim that the world has warmed around 3 degrees Fahrenheit since pre-industrial times is fakery, accomplished by what’s called “the urban heat island effect”. Weather monitoring stations have mostly gradually changed from rural to suburban or from suburban to metropolitan (through construction of many buildings). If the record of only stations which haven’t become surrounded by new buildings was used, there’s been no consistent change in temperature over 100+ years.
The entire narrative is lies. It’s not a mistake but a long lives and well financed fraud.
Just as is the fake discipline called “virology”.
Same financiers, same organisations and same objectives.
It's an interesting, unique and valuable contribution to a large and growing body of data, scientific evidence and research (published by plain old boring humans!) which casts serious doubt on the man-made 'dangerous' global warming hypothesis, a corpus which, by now, should have driven the final nails into the coffin of 'settled' consensus climate science. But we all know why it lives on as Zombie Science - because of huge Green Blob funding, rampant censorship, and almost 100% media and institutional support.
There's lots of excitement among sceptics who seem to think that the word of Grok3 beta is gospel and unchallengeable and hence, in the words of Robert Malone on X for instance, "the Climate Scam is over". I think this is premature. It's not over until the fat lady sings - and I don't think Grok3 beta identifies as the Fat Lady just yet!