[Article] Harvard physicist's take on the VMS

[Koen G]

It looks like the guy in short time generated with AI 36 articles on different fields that aren't his professional area of expertise.
And he is sponsored by AI company so these articles are more an advertisement of their product than the real science.

I will from now on, on a completely personal and subjective basis, refuse to interact with articles like these in any positive way. It is not the reader's task to find any value in your wall of maybe-slop-maybe-not. That is your job as the author. Tell us what you researched, how you did it and what you learned, in a way people can understand. 

As someone who strongly believes in the maxim that any published writing should have first of all the reader in mind, I hate this evolution.

[Jacob Carolus]

Unfortunately, I haven't quite grasped the extent of human involvement or how to use his "Bootloops" tool.

 This is an article about their project. 

You are not allowed to view links. Register or Login to view.

[eggyk]

From this page: You are not allowed to view links. Register or Login to view.

If all you want is the results from the applications described here, the code for those is stored on bootloops.ai (this site). Simply ask your LLM to pull the code or paper you need from here. It's worth emphasizing that the papers here were almost entirely written by Claude, and they are not written well. To save yourself frustration, a recommended approach is to ask your LLM to read them and to explain to you what is in them. Interrogating the science that way may be more rewarding. On the other hand, the papers are actual self-contained science papers with broad introductions, details, plots, discussion, methodology, etc. So they can be read. Just don't expect Shakespeare (except the quotes in the Shakespeare paper).

(Emphasis mine)

Ah yes, let's spam publish long, hard-to-read scientific papers that are almost impossible to check by humans, then imply that we published them to give them credibility. That way, everyone will have to rely on AI to check scientific papers. It's almost like this guy is sponsored by Anthropic or something....    

[Jorge_Stolfi]

Would this work?
You are not allowed to view links. Register or Login to view.

Thanks for the link! But do you mean that the VMS has been compared statistically to that Chinese text?

All the best, --stolfi.

[DG97EEB]

Would this work?
You are not allowed to view links. Register or Login to view.

Thanks for the link! But do you mean that the VMS has been compared statistically to that Chinese text?

All the best, --stolfi.

For example...

Btw, see my note on the Chinese thread.. there's every digitised copy of the Bencao that China knows of, but it's on a Baidu drive I can't access but maybe our Chinese contributor can..

[Ruby Novacna]

Unfortunately, I haven't quite grasped the extent of human involvement or how to use his "Bootloops" tool.

 This is an article about their project. 
You are not allowed to view links. Register or Login to view.

I read that article yesterday; all those files on GitHub are beyond my level of understanding. 
It's great for researchers who are used to working with AI.

[JoJo_Jost]

At some point, the AI will come up with a solution that coherently deciphers the entire text—in a medieval language, with compelling logic, and using a cipher that works—but what no one realizes is that it isn’t the original text; it’s just a text that fits - nothing more.

And no one will notice—or be able to disprove it.

[LisaFaginDavis]

I have more than 200 "solutions" in my inbox from this year alone, nearly one per day! Most are AI-assisted (if not entirely AI-generated). Is one of them "it"? Unlikely - they are generally techno-babble with an acronym to sound official but are clearly AI nonsense.

[npcompl33t]

If AI can 'solve' a millennium problem , i don't think its unreasonable to think it could solve voynich.  

However the millennium problems took nearly $20,000,000 in compute, took approaches humans had already identified as likely, and then pushed them across the finish line with brute force. 

My guess is if you dumped enough $ into compute, and gave it ideas discussed over the years in this forum as a starting point, AI as it exists today might already be good enough. The amount of $ needed probably decreases with every model release.

[DG97EEB]

If AI can 'solve' a millennium problem , i don't think its unreasonable to think it could solve voynich.  

However the millennium problems took nearly $20,000,000 in compute, took approaches humans had already identified as likely, and then pushed them across the finish line with brute force. 

My guess is if you dumped enough $ into compute, and gave it ideas discussed over the years in this forum as a starting point, AI as it exists today might already be good enough. The amount of $ needed probably decreases with every model release.

I'm literally doing that.. you can have access to the Voynich Archive via MCP on my signature.. hook it up to fable 5.5 or Astra and have fun... But... It's not gonna solve it...I don't believe it's a computational problem in the end.. I've even tried quantum (you can stand up a quantum computer on AWS) ... What it is very good for however is research.. plug in Exa and Consensus and the archive which covers everything I can find about Voynich including this forum live updated, and it will spot patterns...and will help you pull threads.. but the stats? So far a dead end like everyone else ...