Will The Real Robot Please Stand Up

AI, blockchain, and the value of being human with Singularity University’s Reese Jones and Project Shivom’s Henry Ines at Blockchain Economic Forum [Interview]

Martine Paris / Via Martine Paris

What does the world’s first decentralized AI humanoid, Sophia the Robot, have to do with owning the data that makes us human?

“A lot!” said Henry Ines, Chief Innovation Officer of Shivom, in his interview with me at the recent Blockchain Economic Forum in San Francisco. Ines, a former DFJ DragonFund partner, just completed a $35 million ICO to fund the building of Shivom’s genomics hub and is now partnering with the makers of Sophia’s brain, SingularityNET, to use AI-driven analytics and machine learning tools to power the dawn of precision medicine.

“A Tylenol that’s right for you might be different than a Tylenol that’s right for me, that’s why the future of medicine is one that can be customized specifically for the biology of individual patients. To get there, researchers need to mine vast data sets. This is why Big Pharma pays millions of dollars each year to companies like 23andMe, to access the data that people donate when they mail in their saliva swabs to find out about their ancestry.”

“At the same time, there are low income, indigenous populations that have zero social capital yet tremendous intrinsic value when it comes to their own genetic makeup. Their genes may hold the key to solving some diseases out there, so we have come up with a way for them to own and profit from their genetic data by logging it onto the blockchain ledger and passing payment directly onto the donor.”

I asked how Shivom is planning for mass adoption when most people don’t yet understand blockchain, particularly in lesser developed countries where some don’t even have a device to manage wallets and keys.

“Yes, UI/UX is very important. We’re trying to make our platform as simple as possible as most patients are pre-occupied with their wellness, not blockchain. Users can buy a physical kit with fiat or simply upload their 23andMe report. Their data is encrypted, anonymized and stored on the backend, and the only thing stored onchain is their HashID, GlobalID and any transactions protected by cryptokeys. Users have complete control of who can access and license their data, and payment might be made on a debit card with tokens which can be redeemed for fiat or healthcare services. For places like India, we’re still working out the model.”

Blockchain technology allowing us to own and profit by our own data is a huge paradigm shift. Most people are so used to giving it away that one can only imagine the cultural ramifications resulting from biological data becoming part of the gig economy.

Martine Paris / Via Martine Paris

I had a chance to speak further about this with Reese Jones, Associate Founder and Board Trustee at Singularity University who was keynoting the conference on the evolution life and data and the role of AI and blockchain.

Reflecting on Sophia, he shared with me his insights on the coming singularity, the point in time in which computers are expected to become smarter than people and runaway technological growth will make things unpredictable with unimaginable consequences.

“We are learning more of what it means to be human, and respond in a human-like way everyday, and technology is getting better at representing itself as sentient, but just when and how AI will ever become a species risk is hotly debated. At the Long Now Foundation, there is a Long Bet between Ray Kurzweil, co-founder of Singularity University, and Mitch Kapor, co-founder of the Electronic Frontier Foundation, that by 2029 no computer or machine intelligence will have passed the Turing Test.”

If you’ve seen Bladerunner then you know that the Turing Test is an interview in which a subject is asked a series of questions to determine whether it is human. Kurzweil who is the mastermind behind deep learning at Google believes strongly that advances in algorithms that drive data acquisition and analysis will be enough to pass the test, but Kapor believes the computer will never be able to fake tacit human experiences that are nuanced and cannot be explicitly expressed or written down.

For anyone who watched Google Assistant call a restaurant at Google IO 2018 back in May, the conversational AI was pretty impressive, certainly had the intonation of a human voice. Not sure where I’d place my bet but certainly looking forward to meeting Sophia in her next iteration.