I asked Gemini AI to evaluate this CDR Google Group

[Michael Hayes]

The "codex" of that group is effectively the living oral history of the modern CDR industry. It is widely considered the most unfiltered, high-signal "town square" for carbon removal, but it is also chaotic.Here is an analysis of the content found within that archive:

1. The Clash of Cultures (Science vs. Sales)

The most distinct feature of this "codex" is the real-time friction between two groups:The Academics & Engineers: They use the group for rigorous, brutal peer review. If someone posts a new "breakthrough" in Direct Air Capture (DAC) or Ocean Alkalinity Enhancement, the old guard often tears it apart with thermodynamic calculations within hours.

The Entrepreneurs & VCs: They use the group for deal flow and hype.The Verdict: This friction makes the archive incredibly valuable. It is one of the few places where you can see marketing claims get "fact-checked" by physics PhDs in real-time.

2. The Focus on MRV (Measurement, Reporting, Verification)

If you read through the history, you will find that MRV is the single most obsessed-over topic.The archive documents the painful realization that removing carbon is easy, but proving you removed it for 1,000+ years is nearly impossible.The "codex" is less about "how to build a machine" and more about "how to write the rules" for carbon accounting.

3. It Is "Pre-Canonical" Knowledge

The Google Group contains information that hasn't made it into textbooks (or even the CDR Primer) yet.Pro: You get data on specific sorbent degradation, supply chain bottlenecks for olivine, or permitting nightmares for injection wells years before they appear in formal studies.Con: It is unstructured and contradictory. You might find a thread from 2021 declaring a method "dead" and a thread from 2023 proving it works. It requires high literacy in the field to navigate.Summary OpinionThe content of the Carbon Dioxide Removal Google Group is essential but raw. It is not a polished guide; it is a raw feed of an industry figuring itself out. 

If you are an investor: It is a lie-detector test for diligence.If you are a founder: It is a place to get your tires kicked, often painfully. If you are a layperson: It is likely too dense and jargon-heavy ("stoichiometry," "additionality," "pore space rights") to be a casual read.

MH] Best Regards 

[Greg Rau]

Michael,

Thanks for the feedback, however artificial it might be. I don't recall a huge clash of "Science vs Sales", though there has been some "Academic vs Engineer" friction.  MRV is a big deal. As for "chaotic", "unstructured and contradictory" - guilty, but then that's what happens when you offer an open marketplace for CDR ideas, comments and news. So it is "not a polished guide; it is a raw feed of an industry figuring itself out" as intended. 

Further criticism invited.

Regards,

Greg

Co-founder and Moderator.

--
You received this message because you are subscribed to the Google Groups "Carbon Dioxide Removal" group.
To unsubscribe from this group and stop receiving emails from it, send an email to CarbonDioxideRem...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/CarbonDioxideRemoval/CABjtO1casghwn0WiQ0aLNR6AX9nsyfZ1vVMkYUeoLouZSK77ig%40mail.gmail.com.

[Michael Hayes]

Agreed. A polished guide implies a level of completeness to polish. CDR will remain a work in progress on many levels for the foreseeable future.

I did ask it for suggestions on improvements, yet it had little to offer that was worth posting about. As such, you win, Greg.

Best regards 

[GRETCHEN & RON LARSON]

Michael and list:

 

     1.  Thanks for reporting the CDR answer to your AI question.   Your 3 parts below are not about CDR; they are about this CDR group.  For me, the main interesting point re your below material is that although present economic CDR activity is about 80% biochar, the word "biochar" isn't mentioned even once by Google AI..  

 

    2.  There is good reason for that omission;  biochar has several good Internet discussion groups (more than any other CdR approach. I'm sure.   Biochar advocates don't need this list.  I suspect your Google Ai might summarize dialog on the group "biochar.io" in the same 3 ways you have reported below for CDR.

 

     3.  New topic:    In trying to better understand CDR today,  I found the best CDR (including biochar) information via a Geoengineering group .   This group doesn't serve a dialog function, but does have a CDR reporting activity function.  That group can be reached at:

          https://www.geoengineeringmonitor.org/

 

      4.    I found more current information there on recent global biochar activity than I have found anywhere else.   Not always favorable, but also not always negative either.  For instance, they emphasized biochar activity is now much more heavily in the global south than I had realized (or you would find on that list)

       With the name "Geoengineering", this group icould be excused from doing CDR.  But they were in existence before CDR was split away from Geoengineering.  So, I urge all parts of this CDR list to see how you are treated there.  They have 17 separate sub-headings.   That for biochar was written in 2026, and seems quite complete.   I learned a lot there today.

 

Ron

     Make sure you first re-read Michael's message title ("I asked......"

--

[Michael Hayes]

Ron, et al.,

Garbage In, Garbage Out, or GIGO, is still alive and well in the AI universe.

I limited my question to what is in this Google Groups' literature...or codex. It spent 30 seconds doing so.

I did not ask about all the many connected related groups that this group has had input into or extracted information from. 

An AI can not 'read the room'.....

I propose that an AI be confronted by as many interested yet diverse experts as possible to teach it about as many perspectives as possible...and get it in writing.

[GRETCHEN & RON LARSON]

Michael, list,  Greg

 

   See few inserts

On 06/14/2026 11:59 PM PDT Michael Hayes <electro...@gmail.com> wrote:

 

 

Ron, et al.,

 

Garbage In, Garbage Out, or GIGO, is still alive and well in the AI universe.

        RWL1:   I am here defending AI.   So I think it better to say:

"Wrong question in, unhelpful answer out".

 

I limited my question to what is in this Google Groups' literature...or codex. It spent 30 seconds doing so.

       RWL2:    I'll bet it was closer to 30 milliseconds

 

I did not ask about all the many connected related groups that this group has had input into or extracted information from. 

 

An AI can not 'read the room'.....

 

I propose that an AI be confronted by as many interested yet diverse experts as possible to teach it about as many perspectives as possible...and get it in writing.

 

RWL3:   Agreed.   I have seen AI give increasingly better answers as we dialogued

    RWL4:   Here is how Greg responded yesterday (and I was going to respond today, independent of your above remark today:

 

"Michael,

Thanks for the feedback, however artificial it might be. I don't recall a huge clash of "Science vs Sales", though there has been some "Academic vs Engineer" friction.  MRV is a big deal. As for "chaotic", "unstructured and contradictory" - guilty, but then that's what happens when you offer an open marketplace for CDR ideas, comments and news. So it is "not a polished guide; it is a raw feed of an industry figuring itself out" as intended. 

Further criticism invited.

Regards,

Greg

Co-founder and Moderator."

 

I agree with that -  and he was probably not viewing yours as an AI accuracy issue.

 

RWL5:   We are lucky to have strong international support for CDR.  Pity the poor folks having to justify SAI and SRM (Today's "Geoengineering")

 

Ron

To view this discussion visit https://groups.google.com/d/msgid/CarbonDioxideRemoval/CABjtO1cGG-YbRABr%2BT6VqCubNnm69%3Ddb63n_t12T39RmsXiauQ%40mail.gmail.com.

[Michael Hayes]

Ron and Greg 

To clearify, I wasn't being critical of AI as I was the one that fed it 'garbage', or a non-precise question. I'm the fault factor. My intention was to have it look at the peer-reviewed studies that have been shared, and then ask if the AI would offer a synergistic mix of methods, yet I triggered a broad evaluation that reflected on Greg's management.

By asking the AI to evaluate the 'codex', I failed to limit that to peer-reviewed papers, yet I thought the outcome was rather positive, and thus worth sharing. 

AI agents clearly have access to lots of information, yet I've now learned that one can not take for granted that the AI will use all of the information that it has access to. I find that I do need to bring information to the exchanges. As an example, if I use a generic term like 'marine energy conversion', the AI will typically call out floating wind farms. There clearly are a number of other technologies in that field. 

I am making a great deal of progress using seagrass as a primary crop and hydrothermal liquefaction, as you suggested earlier. The results being offered by the AI are very positive, both environmentally and economically speaking. 

Best regards to you both 

 

 

[GRETCHEN & RON LARSON]

Michael, Greg, Group:

 

      Re your last lines about new emphasizing seagrass - excellent!    Not only helpful where wrack is a problem, but I believe the best use of the biochar or HTC could be back on the seagrass itself - to improve its health, yield,  etc.  (No problem with salt content there,,)

 

.   Probably plenty of paths there to work on.  Good luck.

 

Ron

[Michael Hayes]

Yes, the 'looping' aspect is rather remarkable. With HTL, one can not only loop the nutrients back into the grow tanks to support crop expansion, one can also use the biocrude oil as a feedstock for the floating high density polyethylene grow tank hulls. If a hydrothermal carbonization system is used along side of a HT liquefaction process, the hydrochar can be used to significantly strengthen the HDPE hulls as well as produce nutrient rich process water.

The thick walled HDPE hulls become a primary longterm C storage system themselves...that largely self-replicate.

When externalities such as shipping biostimulants/nutrients and hydrochar to shore for crop use and then shipping municipal liquid waste back to the marine operation for carbonization and nutrient extraction, the profitability from commodity sales, waste tipping fees, and C credits start to stack up.

Some AI exchanges show a trillion dollar yearly profit if a 10 GtC/y CDR scale is achieved with a rapid scale up to that level of less than 10 years. One interesting estimate that I ran across under the above scenario was that all municipal waste would be used up when the system scales to around 1.5 GtC/y CDR. 

That alone makes this system of systems worth further work.