Matthew Griffin, award winning Futurist working between the dates of 2020 and 2070, is described as “The Adviser behind the Advisers” and a “Young Kurzweil.” Regularly featured in the global press, including BBC, CNBC, Discovery and RT, Matthew’s ability to identify, track, and explain the impacts of hundreds of revolutionary emerging technologies on global culture, industry and society, is unparalleled. Recognised for the past six years as one of the world’s foremost futurists, innovation and strategy experts Matthew is an international speaker who helps governments, investors, multi-nationals and regulators around the world envision, build and lead an inclusive, sustainable future. A rare talent Matthew sits on several boards and his recent work includes mentoring Lunar XPrize teams, building the first generation of biological computers and re-envisioning global education with the G20, and helping the world’s largest manufacturers ideate the next 20 years of intelligent devices and machines. Matthew's clients include three Prime Ministers and several governments, including the G7, Accenture, Bain & Co, BCG, BOA, Blackrock, Bentley, Credit Suisse, Dell EMC, Dentons, Deloitte, Du Pont, E&Y, HPE, Huawei, JPMorgan Chase, KPMG, McKinsey, PWC, Qualcomm, SAP, Samsung, Sopra Steria, UBS, and many more.
WHY THIS MATTERS IN BRIEF
Until now the financial services industry has never been able to benefit from a network effect, but that’s about to change and it could tip Wall Street on its head.
The “Network Effect” is a phenomenon known the world over, it’s where a service becomes more valuable when more people use it.
Over the past decade we’ve seen how technology driven networks replace bureaucratic hierarchies and I think that we can argue, easily, that we are just at the very beginning of this trend. In today’s world the people who create content are often the same people who are consuming the content – you are a node on the network, and, as we hurtle towards the future that’s really important.
That said though, there are no network effects in finance. And you’re not the only one who thinks that’s strange…
In 2008 Satoshi Nakamoto created a game changing new protocol called bitcoin, built on an even more interesting technology called blockchain, that helped create amazingly powerful incentive structures that caused a huge network of miners to come together and provide value. Anyone who holds bitcoin has a strong incentive to make the network better, and as a result, today, the hash power of the bitcoin computing network is greater than the top 100 supercomputers combined.
Now, some entrepreneurs think that it’s time that cryptocurrencies help to bring those same powerful network effects to finance where you have massive hedge funds that are fighting against each other, and where no one really works together. Finance, after all, is still very competitive, very adversarial and there’s no collaboration. And their hope is that they can turn this negative competition space into one that’s highly valuable and highly collaborative.
Richard Craib is one of those entrepreneurs. A 30 year old entrepreneur from South African he runs a hedge fund in San Francisco, and if that wasn’t strange enough for a start then consider this, he doesn’t actually run it.
Confused? Then you’re not alone because as it turns out he doesn’t run it – it’s run, and operated, entirely by an artificial intelligence (AI) that’s continuously being evolved by over 12,000 anonymous data scientists who he’s never laid eyes on.
Welcome to the new era, the era of distributed, autonomous AI operated companies that strangers built.
Craib’s company Numerai, if you can call it his – I’m sure we could debate that – is built on top of a platform that masks the funds trading data before it gets shared with the company’s vast community of anonymous data scientists. Craib uses a form of homomorphic encryption, a relatively niche type of encryption, that lets the scientists run calculations on ciphertext, and create new, better, machine learning financial securities models without letting them see details of the company’s proprietary trades.
“We give away all our data,” says Craib, who studied mathematics at Cornell University in New York before going to work for an asset management firm in South Africa, “but we convert it into this abstract form where people can build machine learning models for the data without really knowing what they’re doing.”
Craib doesn’t know the data scientists because he recruits them online and pays them for their trouble in a digital currency that preserves anonymity.
“Anyone can submit predictions back to us,” he says, “and if they work, we pay them in bitcoin.”
The fact that the scientists work using encrypted data means they can’t use their machine learning models on other data – and neither can he, but despite this Craib believes that the new approach will help him build a better hedge fund, and so far it seems to be working. And he’s not the only one, others, such as Two Sigma are now piling into the space.
Numerai’s fund has been trading stocks for over a year and in that time it’s made money – although strict regulations don’t allow him to say how much – and now more and more big name investors, such as Renaissance Technologies, an enormously successful “Quant” hedge fund, and Union Square Ventures, who just invested $3 million, are ploughing money into the company.
Hedge funds such as Bridgewater Associates, Sentient Technologies and Aidyia have been with algorithms for years now, in some cases decades but Numerai is the first time a founder has combined bitcoin, crowd sourcing, encryption and machine learning, or “AI,” to create one. Even one of its investors, Union Square partner Andy Weissman, calls it an “experiment.”
According to Craib he came up with the idea for Numerai when he was working for a financial services company in South Africa where he was asked to help build a machine learning system that could help run the company’s funds. After working on the project for a while he wanted to share the company’s data with a friend, an expert in machine learning who was working with neural nets, and the company forbade it. It’s that that gave him the initial idea.
“That’s when I started looking into these new ways of encrypting data – looking for a way of sharing the data with him without him being able to steal it and start his own hedge fund,” he says.
Numerai was the result and Craib says he invested over a million dollars of his own money into the venture, topping it up soon afterwards with a $1.5 million injection of cash from a group that included Howard Morgan, one of the founders of Renaissance Technologies, who’s since invested again in Numerai’s latest Series A round, alongside Union Square and First Round Capital.
Everything about the company is unorthodox, and that point’s made even clearer when you watch the short introductory video on Numerai’s website that cuts Craib into a scene that’s more reminiscent of The Matrix.
“When we saw those videos, we thought: ‘this guy thinks differently,’” says Weissman. As Weissman admits, the question is whether the scheme will work. The trouble with homomorphic encryption is that it can significantly slow down data analysis tasks.
“Homomorphic encryption requires a tremendous about of computation time,” says Ameesh Divatia, the CEO of Baffle, a company that’s building encryption similar to what Craib describes, “how do you get it to run inside a business decision window?”
Craib says he’s managed to solve the speed problem but Divatia suggests that this may come at the expense of data privacy. And according to Raphael Bost, a PhD student at Université de Rennes in France who’s also explored the use of machine learning with encrypted data, Numerai’s likely using a method similar to the one used by Microsoft, where the data is encrypted but not in a completely secure way.
“You have to be very careful with side-channels on the algorithm that you are running,” he says of anyone who uses this method.
So far Numerai’s scientists have helped to build over 500,000 machine learning models that have driven over 28 billion predictions for the fund, and as they compete to build the best models they can earn money in the process. In order for the system to really hum though everything has to be done at high volume, and using a machine learning technique called ensemble learning Numerai‘s platform combines the best algorithms together to form a single, more powerful one. In essence the hedge fund’s “brain.”
Every week, a hundred of the scientists earn bitcoin, and so far Numerai has awarded over $150,000 in payouts. If the fund reaches a billion dollars under management, Craib says it’ll end up paying out over $1 million a month to its scientists – and that’s not pocket change.
And as for dealing with encrypted data? Well the scientists involved say it’s like turning the sound off at a party – you can’t listen in on other people’s conversations any longer but you can still get a very good indication on how close they feel to one another.
As Numerai grows, and as it continues to prove that its model works, then who knows, Wall Street might start to turn to the light and embrace collaboration, and then Craib might finally see his Silicon Valley start up dream, the democratisation of finance, come true.