A Cipher Thought Experiment

[oshfdk]

GTMOLRR/O MTNO/S NO/S LO TMO/ST OMP 4TO 4POLO/SNOLPNOLRNO/TGS OMP 4TNO/S KOLTMO NO KLONR LOR/RMOLOTGS SMOO GT/OGK MTNO/S KRMOLO/TMO/S SOMRO | NPMO/RGS LR/OGR GK/ORLT/S NO LOMR ONRLRGS LT/ONOGR MTNO/S MP/OR MPGONR 4O GT/OPMO GS/OGO TRMOLO/T MTNO/S LO/S MPGONR+ | 4OMP GS/OGO 4TLONPMO LOR/RMOLOTGS LPMOONR LOPRMO TO KRLO/KK MTNO/S KOLTMO MTONR NO LOMR LO MP/OOLR KOMR/PLO/RMOLT TO MTO | KLRMONPMO/R SKRNOLPMO GKT/O GK/ROTMO MTO NP/OGSNRO/TMS MRLONRGO/SKRNO/PT MTLO/T GK/OGOLRTNR/T SGO/RPRNO/SMO 4RO | MTO/RMO LO/RMO SMRLO/RT PMO/OPLRMO GTMO/RMO

I will come back with the solution in a few days if no one solves it. I won't indicate if someone is on the right track. But if someone is able to crack it, I'll respond. I hope someone solves it!

BTW, the original message goes (with a few possible corrections, I'm not sure if those are typos or my misinterpretation of the details of the cipher):

======
Hello, this is a test of the feasibility of this code.
I can already see how this creates some very long words.
I am only doing this for fun.
I hope you treat this as fun!

If you have already been able to crack this code, then I am a fool compared to the
clever scribe who wrote the Voynich manuscript.
That wouldn’t surprise me —
there are smart people here.
======

Thank you for this exercise, I learned a few valuable lessons when it comes to using AI for solving various ciphers.

Also, in general this is important in the context of the Voynich Manuscript, because whatever the cipher there is (if it is a cipher), it's very likely that in the course of solving it, after unravelling LAAFU elements and combinatorial issues, one will end up essentially with a multi character substitution cipher as the last step. It's good to know those are not particularly hard to solve given long enough text.

As a side note, I didn't use the image you attached for guiding the AI, but this was a nice confirmation when the Parthenon popped up in the decoded text.