Joining me today is Dr. Marsha Lipton, Co-Founder and CEO of Numeraire Future Trends, a pioneering company that creates a biometric identity layer for high-value physical assets. Marsha, welcome to the show.
Thank you so much, Rachel. So glad to be here.
AI is making documents, images, and provenance narratives increasingly easy to fabricate. How should we think about authenticity?
Even before AI came to the fore, we had a problem which existed before. We have more information than humanity ever had at our fingertips — but I am not convinced that all this information brings us closer to truth. We are increasingly confusing access to information with knowledge, and knowledge with truth. You might have a painting in front of you and the whole body of research, expert opinions, and provenance records — but you are still not 100% sure that all of this information really relates to the object in front of you. And now AI blurs these boundaries even further. It allows the creation of extraordinarily convincing images, narratives, and documentation — so convincingly that it leaves us almost no chance to figure out what is real, what is inferred, and what is just generated.
What are you doing at Numeraire to solve that problem?
In a paradoxical sense, the more powerful the digital world becomes, the more important it is to link it back to the physical world. That is the philosophical underpinning of what Numeraire Future Trends does. We link a body of information — documentation, provenance records — to the physical piece itself. And by immutably linking them, we create what we call forward-looking provenance. It always stays with the object. It develops as it goes along. Certificates, as a piece of paper, exist completely separately from the object — so they can be forged, assembled, separated, or compromised. We link them through the biometrics of the object itself.
What does object biometrics actually mean?
The notion is not foreign to us — all of us have this in our passports. When you go through border control, your passport gives the government permission to access your biometric database. The camera sees how you look now and compares it with the benchmarks stored. If you have identical twins who look the same in a photograph, their biometrics are different. That is what separates them. We do the same for objects. What we capture is not the image of the artwork — it is the structure of the support paper the work is printed on, the fibres of that paper, the blob of paint on an oil painting, the fine scratches on a sculpture's surface, the veining of marble. Those surface imperfections are always unique to each object — and that is exactly what becomes part of the biometric that allows us to distinguish this object from anything else that might look exactly the same to the naked eye.
What does this mean for auction houses and collectors?
It is very important to understand that we are not trying to replace professional authenticators — because there is a big difference between identification and authentication. If you bring us an artwork and say this is a Monet, we cannot miraculously tell you whether that is true. You still need the professional experts in the workflows of auction houses and museums to give their opinion. What we do is make their opinion ever more important — we digitally embed that opinion and it becomes part of the passport. But you can only do it once, at the beginning. Once we link them together, you do not need to repeat it. It is just enough to check that it is the same object which was originally authenticated.
Thank you so much for coming on the show.
Thank you. And I am not alone — we have a really good team. My co-founder is Dr. Tomas Cardona of MIT, who is also CTO of MIT Connection Science. And we have a very strong team of advisors — including a formal physicist from Rockefeller University, CERN, and Argonne National Lab who is leading our AI efforts.
