ARTIFICIAL INTELLIGENCE: Proof that we have been training AI fakes to stab us in the back

In the 1933 film Duck Soup, actor Chico Marx is famously known to have asked, "who ya gonna believe, me or your own eyes?" Fairly meaningless in the 30s, but today, it's more relevant than ever. Let us explain. We know how the ever-expanding capacities of computing power and algorithm efficiency are leading to some pretty wacky technology in the realm of computer vision. Deepfakes are one of the more terrifying outcomes of this. A deepfake can be described as a fraudulent copy of an authentic image, video, or sound clip, which is manipulated to create an erroneous interpretation of the events captures by the authentic media format. The word 'deep' typically refers to the 'deep learning' capability of the artificially intelligent algorithm trained to manifest the most realistic version of the faked media. Real-world applications being: Former US president Barack Obama saying some outlandish things, Facebook founder Mark Zuckerberg admitting to the privacy failings of the social media platform and promoting an art installation, and Speaker of the US House of Representatives Nancy Pelosi made to look incompetent and unfit for office.

Videos like these aren’t proof, of course, that deepfakes are going to destroy our notion of truth and evidence. But it does show that these concerns are not just theoretical, and that this technology — like any other — is slowly going to be adapted by malicious actors. Put another way, we usually tend to think that perception — the evidence of your senses (sight, smell, taste etc.) — provides pretty strong justification of reality. If something is seen with our own eyes, we normally tend to believe it i.e., a photograph. By comparison, third-party claims of senses — which philosophers call “testimony” — provide some justification, but sometimes not quite as much as perception i.e. a painting of a scene. In reality, we know your senses can be deceptive, but that’s less likely than other people (malicious actors) deceiving you.

What we saw last week took this to a whole new level. A potential spy has infiltrated some significant Washington-based political networks found on social network LinkedIn, using an AI-generated profile picture to fool existing members of these networks. Katie Jones was the alias used to connect with a number of policy experts, including a US senator’s aide, a deputy assistant secretary of state, and Paul Winfree, an economist currently being considered for a seat on the Federal Reserve. Although there's evidence to suggest that LinkedIn has been a hotbed for large-scale low-risk espionage by the Chinese government, this instance is unique because a generative adversarial network (GAN) -- an AI method popularized by websites like ThisPersonDoesNotExist.com -- was used to create the account's fake picture.

Here's the kicker, these GANs are trained by the mundane administrative tasks we all participate in when using the internet on a day-to-day basis. Don't believe us? Take Google’s human verification service “Captcha” – more often than not you’ve completed one of these at some point. The purpose of these go beyond proving you are not a piece of software that is unable to recognise all the shopfronts in 9 images. For instance: being asked to type out a blurry word could help Googlebooks’ search function with real text in uploaded books, or rewriting skewed numbers could help train Googlestreetview to know the numbers on houses for Googlemaps, or lastly, selecting all the images that have a car in them could train google’s self-driving car company Waymo improve its algorithm to prevent accidents.

The buck doesn't stop with Google either, human-assisted AI is explicitly the modus operandi at Amazon’s Mechanical Turk (MTurk) platform, which rewards humans for assisting with tasks beyond the capability of certain AI algorithms, such as highlighting key words in an email, or rewriting difficult to read numbers from photographs. The name Mechanical Turk stems from an 18th century "automaton" or self-playing master chess player, in fact it was a mechanical illusion using a human buried under the desk of the machine to operate the arms. Clever huh?!

Ever since the financial crisis of 2008, all activity within a regulated financial institution must meet the strict compliance and ethics standards enforced by the regulator of that jurisdiction. To imagine that a tool like LinkedIn with over 500 million members can be used by malicious actors to solicit insider information, or be used as a tool for corporate espionage, should be of grave concern to all financial institutions big and small. What's worse is that neither the actors, nor the AI behind these LinkedIn profiles can be traced and prosecuted for such illicit activity, especially when private or government institutions are able to launch thousands at a time. 


Source: Nancy Pelosi video (via Youtube), Spy AI (via Associated Press), Google Captcha (via Aalto Blogs), Amazon MTurk

ARTIFICIAL INTELLIGENCE: Amazon's new wearable edges us closer to a reality of emotionally manipulative financial institutions

In the past, we have touched on how a specific device that you use for conversational interface interactions will be locally better at understanding you -- rather than some giant squid-like monster AI hosted on Amazon Web Services. But, what if the conversational interface device is the friendly avatar to such a terrifying AI monster that possesses the ability to emotionally manipulate its user? Well, Isaac Asimov eat your heart out, Amazon are reportedly building an Alexa-enabled wearable that is capable of recognizing human emotions. Using an array of microphones, the wrist-worn device can collect data on the wearer's vocal patterns and use machine learning to build models discerning between states of joy, anger, sorrow, sadness, fear, disgust, boredom, and stress. As we know, Amazon are not without their fair share of data privacy concerns, with Bloomberg recently disclosing that a global team of Amazon workers were reviewing audio clips from millions of Alexa devices in an effort to enhance the capability of the assistant. Given this, we can't help but think of this as means to use the knowledge of a wearer’s emotions to recommend products or otherwise tailor responses.

Let's step back for context. Edge computing is the concept that there are lots of unique distributed smart devices scattered throughout our physical world, each needing to communicate with other humans and devices. Two layers of this are very familiar to us: (1) the phone and (2) the home. Apple has become a laggard in artificial intelligence -- behind Google on the phone, and behind Amazon and Google at home -- over the last several years. Further, when looking at core machine learning research, Facebook and Google lead the way. Google's assistant is the smartest and most adaptable, leveraging the company's expertise in search intent to divine meaning. Amazon's Alexa has a lead in physical presence, and thus customer development, as well as its attachment to voice commerce. Facebook is expert in vision and speech, owning the content channels for both (e.g., Instagram, Messenger). We also see (3) the car as developing a warzone for tech companies' data-hungry gadgets.

Looking back at financial services, it's hard to find a large financial technology provider -- save for maybe IBM -- that can compete for human attention or precision of conversation with the big tech firms (not to mention the Chinese techs). We do see many interesting symptoms, like KAI - a conversational AI platform for the finance industry used by the likes of Wells Fargo, JP Morgan, and TD Bank; but barely any compete for a relationship with a human being in their regular life. The US is fertile ground for this stuff, because a regulated moat protects financial data from the tech companies. Which is likely to keep Big Tech away from diving head first into full service banking, but with the recent launch of Apple's AppleCard we are starting to see vulnerabilities in that analogy. So how long can we rely on the narrative so eloquently put by Chris Skinner"the reason Amazon won’t get into full service banking is because dealing with technology is very different to dealing with money; furthermore, dealing with money through technology is very different to dealing with technology through money"? Also, how would you feel about your bank knowing when you are at your most vulnerable?


Source: Bloomberg Article, KAI Platform (via Kasisto)

VIRTUAL REALITY: Enterprise applications of VR prove we are on track for a $200 billion mixed commerce market by 2025

We stand by our position that mixed reality seems to be headed more towards large, enterprise use-cases like city planning, construction, low skilled worker on-site instruction for utilities or manufacturers, and the military. Yet among young consumers, the behavior of buying digital goods in video games, and the associated monetization of content from video games using channels like eSports continues to be a powerful secular trend. Billions of revenue are generated by free games that only sell cosmetic in-game objects. See, as proof points, the fast growth of Twitch users and the $1B+ in revenue Fortnite made from microtransactions. Last week, Facebook doubled down on the former enterprise-centric use case for mixed reality -- announcing its Oculus device-management subscription for enterprise users. The subscription will cost $180 per device per year and promises "a dedicated software suite offering device setup and management tools, enterprise-grade service and support, and a new user experience customized for business use cases" (see here). Evidently, companies deploying mixed reality solutions generally see better customer retention, satisfaction and operating metrics. Take VR surgical training platform OssoVR -- who claim to have witnessed a 230% improvement in performance by surgeons training in VR. Whilst Walmart admits to VR training boosting employee confidence, retention, and overall training test scores by 10-15%. And let's never forget the VR training platform for cooks in fast-food giant KFC's Chicken Mastery program -- the nightmare-sh and BioShock-esque “escape room” replete with narration from an omnipresent, mildly demonic Colonel Sanders. Apart from giving trainees a mild post-traumatic stress disorder, the training platform (on average) reduced instruction time by 60%.

In financial services much of the framework-setting falls to a centralized function, whether that's a Chief Investment Officer creating portfolios or a more decentralised one i.e., branch or advisor office role assisting in the task of consolidating accounts, or discussing mortgage finance options. Yet realistic presence and emotional resonance, via a truly immersive experience, still matter. Facebook Reality Labs, recently announced, that it's working on this -- bringing full-body avatars to its Oculus experience. Will this allow us to emotionally connect with others in a virtual setting or merely remind us that virtual worlds have no place for such complexity? Either way it's important to note that in our latest payments report we estimated the install base for AR/VR active devices to reach 1 billion by 2025, fueling a revenue pool for mixed commerce of $200 billion at the same time. Seemingly, we are on track.


Source: Oculus for Business, KFC Virtual Training Room (Youtube), Facebook Full body VR (via CNN)

ARTIFICIAL INTELLIGENCE: 10,000 People at Citi Fertile for Machine Processing


Nothing quite makes human people like us perk up more than being described as "most fertile for machine processing". And yet, that's exactly what Jamie Forese, the president of Citigroup told the FT about the investment bank's personnel. Out of 20,000 operating roles, he sees 10,000 potentially going away over the next five years. Now that 50% is a pretty big number, and not everyone agrees. HSBC, for example, see only 5-10% more automation potential over the same time period. So let's cut the pizza at 25%, and still gawk in disbelief.

Yet, this live data point is exactly in line with our analysis of what artificial intelligence will do across the financial services sector. In Augmented Finance, we identified $1 trillion of cost at play across banking, investment management and insurance. Behind that cost are those real human people -- 2.5 million of which are in the United States, and about 160,000 of which sit in the middle office of investment and banking organizations. Looking at the last 10 years only, the FT found 60,000 jobs cut from the top investment banks. The way things are headed, sounds like it's time to take programming courses at General Assembly.

None of us should be glib about the potential impact on the lives of employees of these companies. One of the ethical questions with which we struggle is -- whose responsibility is the welfare of these employees? Does Google and free machine learning software share a responsbility? Do CEOs of too-big-to-fail finance firms share a responsbility? Do investors looking for cost cutting share a responsibility? Does the consumer, wanting to pay nothing for banking, share a responsibility? And if yes to all, how do we come together to make a world where we celebrate not the cost cutting potential, but instead the potential human productivity growth? Is technology a shield or a sword?


Source: Autonomous NEXT (Augmented Finance), FT (Citi), Look and Learn (Lamp Lighter),

FINTECH: What Will 2018 Fintech & Crypto Look Like?

We are thinking about 2018, and here you will find our best educated guess about the year to come. 

Crypto Eighteen  

If you thought 2017 was loud about crypto, just wait till 2018. Up or down, that doesn't matter -- what will certainly be in play is massive volatility as the crypto economy beats on against traditional finance, regulators and sovereign power. The largest mountains to climb are the development of institutional crypto custody and a vanilla ETF product to absorb the splurging demand, and we think this will happen. In terms of creative destruction, we expect one of the top ten 2017 currencies to collapse 80%, one of the enterprise blockchain consortia to fall apart, and new technical solutions like the Tangle or Hashgraph to challenge our assumption that Bitcoin is the endgame.

Source: Pexels CC0

Source: Pexels CC0


Augmented Commerce 

Let's go out on a limb, with that limb being a 3D rendered object in virtual reality. We think there's a storm brewing in digital goods spilling out into our real world (think Crypo Kitties), and physical goods becoming virtual (think Ikea). Machine vision combined with Whole Foods, Amazon's augmented reality app, and the iPhone X signals to us that a new type of commerce is emerging. Symptoms like the dominance of eSports and the popularity of sponsored SnapChat filters will only increase, and lead to a new purchasing and payments experiences. Financial companies will miss this completely.

Source: Minecraft

Source: Minecraft


Social Selling Meets Propaganda Bots 

How can financial advisors, insurance agents, bank tellers and other human front office staff compete with bots? How can they compete with Kim Kardashian and kitten GIFs for attention? They can't -- at least not without some automated help. We think that 2018 will see a much fuller implementation of Social Selling, i.e., using social networks like LinkedIn to prospect for business, and that this channel will become plugged into roboadvisors, neobanks and insurtech startups. Further, social selling is all about content marketing, by using writing, podcasts and video. To distribute these at scale, we expect the technology behind propaganda bots to find a way into the mainstream economy and become a more acceptable strategy. Call it demand generation.

Source: Pexels CC0  

Source: Pexels CC0 

Here is the long form update on all the themes we are tracking for next year.

#Blockchain for Enterprise

  • A major incumbent custodian will offer an institutional platform for crypto assets and will see inflows of $5 billion in response
  •  Traditional securities, like equities and fixed income, will be traded on a production blockchain platform that also supports crypto instrument trading (not just futures, but the actual thing)
  •  An industry consortia will fall apart and give up its proprietary blockchain format for a public alternative instead; and a non-blockchain technology, like the IOTA tangle or the Hashgraph, will create a new rush of excitement for incumbents 
  •  1,000 back-office financial services staff will be laid off resulting from settlement / reconciliation automation

#Bitcoin & Initial Coin Offerings

  • The SAFT will face a well respected challenger for how to legally paper crypto currency investing
  •  Five exchange traded funds will be in the American market by year end, two of which will purely track single instruments (Bitcoin, Ethereum) and three of which will be indexes of the top coins. Decentralized exchanges fail to be widely adopted due to a lack of global liquidity.
  •  There will be 500 crypto funds managing 20 billion in assets, using strategies from AI to distressed investing, and two of them will be managing over a billion each.
  •  Protocol-level token interest is replaced and outgrown by use-case/app level token interest (e.g., machine economy, fintech startups) in terms of ICO funding and human capital. ICO funding eclipses all Fintech venture capital funding.
  •  One of the top 10 crypto currencies today will collapse over 80% in market cap


  • Things get worse before they get better. Deep Fakes (i.e., videos with superimposed faces of other people and their voices) spread by propaganda bots hit the Internet and cause a media stir, with a particularly painful media scandal affecting a political party
  •  The Asian artificial intelligence giants (Alibaba, Baidu, Tencent) outcompete anything Google or Facebook can do by (1) creating a more powerful neural network cluster, say by a factor of 2x, (2) investing twice as much capital into R&D, and (3) having unprecedented access to government data
  •  Decentralized Autonomous Organizations with certain functions defined by machine judgment start to battle with the traditional economy


  • An ETF with a price point of 0 bps will be created
  •  Mid-sized stand-alone roboadvisors begin to die in droves, as venture investment into the space moves onto greener pastures. Seed stage wealth tech investment is lower than 2015 numbers. 
  • A micro-investing service reaches 10 million users, introduces crypto currencies and becomes solvent based on the crypto-fees
  •  One of the incumbent roboadvisors (Schwab, BlackRock, et al) turns on industrial-strength content farming and social selling, clogging up LinkedIn and Twitter with bot armies of asset allocation advertising

#Neobanks & Digital Lending

  • Neobanks will become cryptobanks, and cryptobanks will become neobanks. Personal Financial Management and budgeting will recede into the background as currency conversion becomes the most important part of a mobile bank app. The category will raise $2 billion in funding.
  •  The two largest digital lenders by underwriting volume will be Amazon and Goldman Sachs, or similar high tech and high finance players. Standalone digital lenders will continue to struggle with scale and capital.
  •  A neobank will get hacked, and will thereafter die

Financial #APIs and Banks-as-a-Service

  • Open banking and PSD2 will hit Europe like a tsunami and force the opening of data ... but nothing drastic will actually happen. Bank-as-a-service infrastructure startups, like ClearBank, may see a few dozed $ million shift onto their platforms, but incumbents will not see major consumer behavior change.
  •  The data availability will contribute to the progress in virtual financial assistants, who can move money between accounts, spot the top interest rate, and tell you about it on chat and voice channels
  •  An enterprise tech player, like Oracle or IBM, will package all the bank APIs together into a single service offering and capture all the usage

#Chatbots & Voice

  • 3D rendered avatars begin to represent traditional brands in virtual worlds and retail locations, and are able to sell financial products
  •  One of the voice assistant, like Alexa, Google Home or HomePod, will have 100,000 skills, but none of them will be good at financial advice just yet
  • A startup focused on being the virtual assistant for financial services will reach 5 million users

#Regtech, #Crowdfunding

  • Centralized government technology becomes a major investment theme, as sovereigns battle decentralized crypto chaos, North Korea hackers, Russian propaganda bots, and Asian AI giants. A marquee check of over $5 billion is spent on a GovTech venture.
  • Identity solutions -- face ID, blockchain KYC/AML -- create the impression that we are safer. But another massive 100 million person hack happens and catalyzes a major innovation in identity theft recovery (rather than prevention).
  • Billionaire crypto whales becomes patrons of a new wave of art and entertainment, especially in VR and eSports


  • Augmented reality payments are seen in the wild, using for example machine vision on fruits and vegetables at Whole Foods to create an experience without checkout
  • International B2B payments begin to see margin collapse as Ripple, or potentially consortia technology, begins to meaningfully eat into SWIFT's network and technology
  • Standalone messenger apps like Kik will fail to replicate the success of WeChat in combining payments with messaging, but one of the GAFA will see an unexpected surge in payments volume within its broader platform and will offer a savings product


  • Claims assessment using machine vision goes mainstream at a top 5 insurance company, and results in 500 layoffs for claims adjusters
  • A B2C insurtech startup like Lemonade reaches 500,000 customers, with a better loss ratio than its legacy competitors
  • Another unprecedented hurricane season catalyzes large insurers to campaign against climate change and for green energy

#AttentionEconomy & Millennials

  • Millennials are supposed to inherit $30 trillion in a wealth transfer over the next 20 years, but crypto will have accelerated that transfer by 5 years
  • The number of hours humans spend on media will have peaked at 12 hours, and will stay flat in 2018
  • Media companies will begin to optimize not for attention but for emotion, and advertisers will expect to be able to buy an emotion associated with their brand and web traffic
  • The quantification of human attention units, being built by projects like Brave or GazeCoin, will be integrated as a core functionality by one of the GAFA

#VirtualReality & #AugmentedReality

  • Virtual Goods (e.g., 3D rendered objects in video games and virtual worlds) continue to grow in value relative to the overall economy, and a massive commerce company for AR/VR objects is born
  • Augmented Reality social networks and games will be the new normal. A new viral phenomenon, bigger than Pokemon Go, will take the world by storm.
  • The number of VR headsets sold, however, will be a disappointment as dedicated VR content struggles


  • There will be 3 times more smart IoT devices in the world than human beings. A new botnet more powerful than anything we've seen yet will be run off several hundred thousand of these devices to attack Internet services.
  • On-demand risk insurance -- either via telematics inside of smart cars or biometrics / chip implants inside of humans -- will become a major source of innovation, similar to on-demand cloud storage or processing power 
  • The machine economy, powered by AI aggregators like Amazon or Hut34 and cryptocurrencies like IOTA, will begin to take shape, with a marquee $100 million venture investment in the space as a signal