Welcome to one of the most active flamenco sites on the Internet. Guests can read most posts but if you want to participate click here to register.
This site is dedicated to the memory of Paco de Lucía, Ron Mitchell, Guy Williams, Linda Elvira, Philip John Lee, Craig Eros, Ben Woods, David Serva, Tom Blackshear and Sean O'Brien who went ahead of us.
We receive 12,200 visitors a month from 200 countries and 1.7 million page impressions a year. To advertise on this site please contact us.
I queried the AI search mode on Chrome to find answers to various questions about whether AI will eventually be able to build better classical ( flamenco) guitars than an experienced guitar maker.
Posts: 16158
Joined: Dec. 14 2004
From: Washington DC
RE: AI and Experienced luthier - Q and A (in reply to estebanana)
I am not clear on why AI can't facilitate history and context? The other Luthier essentials sound pretty "woo woo" to me.
The missing thing about "feel" is really about "stiffness", again, I can't see why some device used by AI can't determine stiffness the same (or better) as experienced luthier fingers. But the main physical build aspects seem doable by AI, scary. It can already make you cry with hallmark card style fake music about your dead dog.
Once it can do that with it's own 3d printed guitar, it will own us.
Posts: 4720
Joined: Aug. 9 2006
From: Iran (living in Germany)
RE: AI and Experienced luthier - Q and A (in reply to Ricardo)
Chat GPT :
Will an A.I. studio eventually build a flamenco guitar just like an experienced luthier?
Short answer: **yes—very likely**, but with important caveats.
### Why it’s plausible
An “AI studio” (meaning AI + robotics + advanced manufacturing) could eventually match an experienced flamenco luthier because:
1. **Measurement & modeling**
* High-resolution scans (CT, laser, vibration analysis) can capture the geometry, thickness gradients, bracing patterns, and resonant behavior of great flamenco guitars. * Machine learning can correlate these variables with sound characteristics like *attack, dryness, rasgueado response,* and *percussiveness*.
2. **Material understanding**
* AI can statistically analyze cypress, spruce, and cedar—grain orientation, density, stiffness, damping—far beyond what the human eye or tap-testing can perceive. * Over time, it could select or even *engineer* wood (or wood-like composites) that behave consistently like top-tier flamenco tonewood.
3. **Precision fabrication**
* Robotic carving and assembly can achieve tolerances no human can repeat consistently. * AI-guided voicing could adjust thicknesses and braces iteratively based on real-time acoustic feedback.
4. **Learning from masters**
* If trained on enough instruments by great luthiers (old Conde, Barbero, Reyes, etc.), AI could replicate their *average* decision patterns extremely well.
### Where it gets tricky
Flamenco guitars are especially hard because they’re not just about “good tone”:
* **Controlled imperfection** Great flamenco guitars often sound alive *because* they’re pushed to the edge—light builds, fast decay, slight asymmetries. Humans sometimes break rules intuitively; AI tends to optimize.
* **Player-specific voicing** Luthiers subtly adjust for hand strength, golpe style, and repertoire. That kind of embodied empathy is hard to encode.
* **Aesthetic and cultural judgment** Flamenco sound is culturally defined and evolving. AI can imitate preferences, but deciding *what flamenco should sound like next* is harder.
### The most realistic outcome
Rather than replacing luthiers, AI studios will likely:
* Produce **exceptionally consistent, high-quality flamenco guitars** * Match or exceed **90–95%** of what great luthiers achieve * Make elite-level instruments more accessible * Still leave room at the top for human masters whose work is prized for individuality, history, and soul
Think of it like this:
> AI can learn how to build a great flamenco guitar. > A master luthier knows *when* to break what they’ve learned.
### Bottom line
Yes—an AI studio will eventually build flamenco guitars **as good as** those of experienced luthiers in sound and playability. But the very best human-made instruments will likely remain special, not because AI can’t copy them—but because flamenco values **human risk, tradition, and intention**, not just results.
RE: AI and Experienced luthier - Q and A (in reply to Ricardo)
quote:
ORIGINAL: Ricardo
I am not clear on why AI can't facilitate history and context? The other Luthier essentials sound pretty "woo woo" to me.
The missing thing about "feel" is really about "stiffness", again, I can't see why some device used by AI can't determine stiffness the same (or better) as experienced luthier fingers. But the main physical build aspects seem doable by AI, scary. It can already make you cry with hallmark card style fake music about your dead dog.
Once it can do that with it's own 3d printed guitar, it will own us.
I read Arash’s questions to Chat GPT and I can see a possibility that AI could make a pretty good guitar with AI guided robotic routers and CNC machines, ect. I think they would be like the mass produced guitars in Valencia’s fabricas. But this would still require humans to select the materials and judge how to give the technology a look at it.
A guy in our Akune Guitar Circle is a retired surgeon who plays guitar. Not only is he a party animal, but he was the head surgeon at a big hospital and trained three decades of surgeons. He told me at the central hospital in Kagoshima city that there are already remote robotic operataing theaters. The doctors sit in a computer/ video suite about 700 meters away on the other side of the hospital and conduct the operations by guiding the robots. He says they have several live video feeds to look at the operation through tiny cameras in delicate tubes in the body and the outside.
I asked him as a conventionally trained surgeon if this is better and more accurate and he said yes it’s quite good. The surgeon still needs to be qualified to operate in a conventional way, but they use the operating suite as a tool.
These suites cost millions of dollars, but theoretically this system could build a guitar. The question is for the robotic instrument directed by a human luthier, when does the guitar building suite become economically viable, understanding that making guitar is about as complex as a thoracic surgery procedure.
Posts: 16158
Joined: Dec. 14 2004
From: Washington DC
RE: AI and Experienced luthier - Q and A (in reply to estebanana)
The big main factor with high end guitars is the label itself. So here is what I think will occur.
Yes we will get terminator guitars very cheap as this system will replace the "fabricas" as you say, perhaps put them all to an end such there is one master output center and costs will only reflect materials and looks. The "label" concert guitars will continue to be made and be collectable. However, at some point, some guitar salon ****s will do exactly as Manuel Ramirez did...start making "fake" torres guitars indistinguishable from the originals. The guitar market will have to then construct "blade runner" type cyborg guitar appraiser police hunter/killers that can do Richard Brune's job of spotting the fakes, trace them to the source build centers and take down the operation. The flamenco guitar robo cops of the future will probably carry hot glue guns and have extreme privileges in society.
Printable Version
New Messages
No New Messages
Hot Topic w/ New Messages
Hot Topic w/o New Messages
Locked w/ New Messages
Locked w/o New Messages
Post New Thread