Amazon is ‘doubling down’ on digital ID credentials, and maybe Worldcoin is on the right track
Plus: Can consumers "own" an AI, the EU’s weird €4M bounty on age verification (not an EU Digital Wallet) and the Empowerment Tech use cases to watch at the DIF hackathon
Hi everyone, thanks for coming back to Customer Futures.
Each week I unpack the disruptive shifts around digital wallets, Personal AI and digital customer relationships.
If you haven’t yet signed up, why not subscribe:
I’ve been thinking.
About the human side of Empowerment Tech. Not just about the customers, consumers or users of ET.
Rather, I’ve been thinking about us.
The Customer Futures network. Those of us actually working on the future of being a digital customer.
Working on ET itself.
The people ideating, designing and building. The people developing, delivering and testing. And the teams inside businesses who see the vast potential of ET, and are making the case for change. The use case, the business case, the investment case.
I’m going to make a prediction.
You feel that the pioneering work on ET is exhilarating, fascinating, and important. But it’s also exhausting, frustrating and requires weapons-grade levels of patience.
Why? Because this market is hard. And it’s supposed to be hard. In every corner of the ET market, it’s almost entirely new.
There isn’t a neat playbook for this.
How to think about wallet-to-wallet identity flows. How to design new commercial models for tech that’s not yet adopted at scale.
How to design the governance for these new data ecosystems. How to overlay existing legal frameworks or even to talk about it with compliance teams.
How to build the new user experience for an AI Agent. How to prove that it’s an AI agent at all.
If it was easy, it would have been done by now. But it got me thinking.
The Customer Futures network is special. It’s full of people paying attention to portable data, to privacy and to consumer empowerment. But also to digital wallets, Personal AI and verifiable credentials.
And it’s full of experts.
People with the professional and emotional scars from working at the coal face of the ET market. Leading, shipping, selling.
People who have successfully raised money for their ET projects, often the same people who have already tried and failed to build a pioneering personal data platform.
People working at large companies who believe ET can make a real difference to millions of customers, and can drive the next generation of digital transformation.
People who have tried eight different ways to talk about digital wallets and credentials with the ‘stakeholders’ who care about ‘ROI’ and ‘customer engagement’ this quarter.
And all the ET specialists in all the different niches - from UI and legal experts, to credential and system designers, to wallet architects and privacy teams.
Here’s where I’m going with this.
Having spent the last 18 months researching it, and speaking with many of you, I can see there’s a disconnect.
A gap between the teams who need some short term expert help, and the people and skills in the network.
Perhaps your ET project
Needs some temporary digital wallet expertise to plan and deliver a development sprint around personal data and SSI
Wants to understand the state of the art on trusted UI design and onboarding experiences
Needs to sense-check a decentralised ID architecture, or get an expert opinion about on mDL vs. VCs
Needs to discuss the investment case for verifiable credentials
Wants to understand how to respond to the latest developments with the EU Digital Identity wallet
Is exploring how Personal AI will impact customer engagement in a new product launch, and wants an outside sense-check
Needs a legal opinion on the T&Cs for a digital wallet (but don’t want to spend time and effort and cash just to educate your lawyer on the new data flows)
So here’s the question.
Can we help?
I’m working on a new way to connect the right ET people to the right ET projects at the right time.
Because there are now over 3000 of us signed up to the Customer Futures newsletter across different channels.
That’s a lot of people interested in ET. Who care about digital customers. And who between us have the skills, expertise and capability to move the market for customer-controlled digital experiences.
We certainly don’t have all the answers. But together we probably know someone that can help.
Do you need short term help with digital wallets, verifiable credentials, data stores, Personal AI, or anything ET… however small?
Get In Touch: experts@customerfutures.com
We can help you find the skills and expertise you need. Maybe it’s an extra developer, tester or project manager for 6 weeks to get through the sprint. Maybe it’s only an expert Zoom call. Or a workshop.
Let us know.
Or are you are one of those experts looking for ways to get connected to a digital wallet project or ET role?
Maybe your contract has just finished. Maybe you’re about to leave your job. Or perhaps you’re available for an expert call on a specific topic or on how ET impacts a sector.
Get In Touch: experts@customerfutures.com
You can also just reply to this email, or DM me here. All conversations will be private and confidential, and under NDA if that’s needed.
Look, it’s early. It’s an experiment. But it’s an opportunity to help each other. So let’s get to it. Working on the future of being a truly digital customer.
And welcome back to the Customer Futures newsletter.
In this week’s edition
Can consumers "own" an AI that makes financial decisions for them?
Empowering Consumers with Personal AI Agents: Legal Foundations and Design Considerations
Wallet wars or digital public infrastructure?
Amazon is ‘doubling down’ on digital ID credentials
Empowerment Tech use cases to keep an eye on at the DIF hackathon
The EU’s weird €4M bounty on age verification (not an EU Digital Wallet)
Maybe Worldcoin is on the right track, but tech choices are getting in the way
Let’s Go.
Can consumers "own" an AI that makes financial decisions for them?
This week’s Fintech long read is required reading from Simon Taylor.
If AI moves to the device, how do we manage risk? And Simon expertly asks: “if a customer can own a car, a house and stock, can they own an AI too?”
Financial Services will be a high watermark for managing the risks around Personal AI. But in what other areas of life, perhaps helping with daily decisions, will people hand over to their Personal AI?
Health? Travel? Home? Media?
We’re about to see an explosion of Personal AI use cases. It’s about to get interesting.
Empowering Consumers with Personal AI Agents: Legal Foundations and Design Considerations
While we are on the topic of AI agents and regulations, here’s one of the most important pieces you’ll read this week.
It’s about the existing legal frameworks and design modes that are going to be critical to the adoption of LLMs on the customer side.
“For example, imagine a personal AI agent negotiating a better price for a subscription service on your behalf. Under UETA, this automated transaction would be legally binding, just as if you had negotiated it yourself.
“Beyond price negotiation, such agents could automatically handle your insurance claims, gather quotes for home repairs, or even help you manage your investments according to your risk tolerance.
“Imagine receiving proactive alerts from your AI agent about better deals on services you frequently use or having it automatically adjust your utility plans based on your actual consumption patterns to save you money. These examples illustrate the potential of personal AI agents to simplify our lives and give us more control over our interactions with complex systems.”
Dazza Greenwood is no slouch here. He lectures on this stuff at MIT, and is a pioneering legal researcher and advisor on AI.
My favourite part of the piece gets into the likely design modes for Personal AI. How we will need to balance security, user experience, and the legal or regulatory constraints. He sees three options:
“A Full Authentication Model, whereby the agent uses the same authentication and authorization credentials as its user;
“An Intermediary Model, whereby the agent is operated by another party who uses the agent to act on the user’s behalf;
“A Decentralized Identity Model, whereby the agent leverages decentralized identifiers and verifiable credentials to interact with third parties, giving users direct control over their digital identity.”
Dazza is putting the lettuce (legal) into the BLT sandwich of ET.
It’s juicy stuff, and gets to the guts of how this is all going to work in practice. Bravo.
Wallet wars or digital public infrastructure?
A timely paper on how governments can roll out digital wallet platforms for citizens. It’s an analysis of government consultations and over 80 papers.
As you might expect, they conclude there are two market models: 1) Government ID-Infrastructure Wallet (basically eGovernment) and 2) Trust ID Wallet Federation (a competitive market).
We’re seeing both play out in the EU. Germany’s SPRIN-D program is an excellent example, where their government ID wallet will be offered alongside a marketplace of private sector ones (and where notably Google is an official member of the innovation track).
Amazon is ‘doubling down’ on digital ID credentials
It turns out that Amazon may well help accelerate the adoption of portable digital IDs.
Like everyone else dealing with today’s hot mess of ID document verification, Amazon’s ID team are “starting to hate” document scanning and traditional ID checks.
As a result they are now leaning in to reusable digital IDs. Approaches like the mobile driver’s license (mDLs) and the EU Digital ID wallet. Apparently they will be accepting them from next year.
It’s worth paying attention when Amazon’s ID lead says, “we are doubling down on digital credentials.”
Empowerment Tech use cases to keep an eye on at the DIF hackathon
The Decentralised Identity Forum (DIF) is in the middle of running some interesting hackathons. Each has a different theme, and is sponsored by different groups and vendors.
Here are a couple of the interesting use cases:
Verifiable Learner/Worker IDs and Records… establish credibility in digital credentials, focusing on how humans can trust verification processes in a world of diverse VCs
eKYC with Reusable Identity… build the Expat Wallet that orchestrates the entire flow using a pre-built To-Do List
KYC for repeat customers every time they transact from different payment apps… store the ‘known customer credential’ in a Decentralized Web Node (DWN) so that she can present it to the business from any payment app
Privacy-preserving Personhood Credentials… with one PHC per person (no duplicates per individual), and ‘unlinkable pseudonymity’ (the user can interact anonymously through a service-specific pseudonym)
Frictionless and connected travel journey… from looking->booking->arriving->enjoying->departing-reminiscing->returning, demonstrating the reusability and portability of traveler information along the journey
It’s worth calling out a couple of themes here from an Empowerment Tech point of view.
And a caution: we need to look at the pattern here, not just the specific use case or vendor.
1. Adding an identity credential to a point of sale terminal
TBD are sponsoring this one. Being able to present ID credentials along with payments, and using different payment apps.
I’ve been writing about ‘Payments Plus ID’ for a while. Especially when you look at the PoS terminal as an immediately valuable credential acceptance network. If the merchant can just ‘switch on’ the Pay+ID software, they could start requesting specific credentials at the checkout (and even online). IS-VIP, IS-OVER-21, IS-LOCAL-RESIDENT.
Hello Stripe and Square.
2. Using verifiable credentials for verifiable AI
Digital consumer tools often start out looking like toys (browsers were chat rooms, and GenAI was ‘make me look old’ apps). Personal AI will do the same thing (“tell me something I didn’t know about myself”) before getting useful.
As AI explodes on the consumer side with AI Agents, we’re going to need to know what-is-what, and who-is-who.
Based on what I can see being developed right now, soon Personal AI will be:
“Analyse my monthly spend, and set up some saving accounts to automatically move money when I hit my monthly limits.”
“Review the family diet and create a new weekly meal plan… then order it to arrive on Sunday.”
It’s going to mess with KYC, online bookings and payments in a major way. As we saw in the posts above about Personal AI regulation, we’re going to need ‘verified AI’, and fast.
Which is why Cheqd are sponsoring this hackathon track to explore what’s possible using their SSI payment network.
I also saw that Okta has started this journey too, with ‘AI credentials’. New authentication flows for GenAI, allowing consumer agents to call APIs of services like Google Calendar, Github and Box on the user's behalf. Critically, it’s about knowing which AI is connected to which user.
Keep an eye on this one.
DIF HACK PROJECTS, VERIFIABLE AI, OKTA AI CREDS
EU offers €4M to solve online age verification using verifiable credentials - why not use the eIDAS2 wallet?
This is an odd one. On one hand, the EU is rolling out Digital ID wallets across the entire bloc of nearly 30 countries.
Which includes the messy work of defining wallet architectures, credentials standards, and harmonising the tech. Not to mention managing the politics, the patchwork of existing regulations and needs, and of course lobbying by vendors.
On the other hand, other parts of the EU have kicked off parallel programs to solve ID-related use cases. Like an age verification, into which the EU commission has offered to pump €4M for a working solution.
Here’s the latest announcement about an age-verification bounty:
“This procurement will enable users of online web services to prove their age through the presentation of an electronic attestation through a dedicated application (mobile app) in a privacy preserving manner in order to access age restricted content.
“This mobile app will directly contribute to the implementation of the provisions on protection of minors of the Digital Services Act (DSA). The application will be built in compliance with the emerging technical specifications of the European Digital Identity Wallet”
Four million euros, no less. For consulting, designing and building another EU app that can do age verification.
It will be related to, compliant with, but different from… the EUDIW.
It seems there’s also an ‘EU Digital Travel app’ being developed by the EU, centrally. Again, related to eIDAS2, but separate. More confusion.
As Andy Tobin, eIDAS2 expert points out:
“1) Why create/commission new separate apps when millions of Euro are being spent on eIDAS. Won't these two new apps muddy the waters for eIDAS and completely confuse the public? Why not just use eIDAS wallets.
“2) Why the focus on mDL/mDOC (notably supported by Apple and Google already)? What's going on there?
“3) Is this an acknowledgement that eIDAS wallets won't be ready to support the Digital Services Act, and a stop-gap is needed?
“4) If the Commission is happy to "centralise" the development of two apps of such importance and allow their use across the EU, why not do the same for eIDAS and initially develop a single EU-wide eIDAS wallet, with one certification, and a nice toolkit for issuers/verifiers to use. This would get eIDAS off the ground far more quickly and cohesively, reducing the work required by member states and the private sector. Later on, certification could be opened up for other apps, but start simple and get things moving.”
The EU has 27 Member States and over 300M citizens. You can forgive them for the left hand not talking to the right.
But these other ID app efforts risk confusing users and businesses. And frankly the experts in the room who are trying to navigate an already complex eIDAS2 roll out.
We can only assume there will be more mini projects like this. Where other apps will be ‘compliant with’ the EU Digital ID Wallet, but not use it directly.
Maybe that’s a good thing? Maybe it will drive adoption for single-use ID apps (one hammer, one nail)? Or maybe it will make the market even more messy, where citizens and customers have multiple apps to use for the same thing? I guess that’s what happens with payments today.
If a supermarket wants to check someone is Over-18 to buy alcohol, and the individual wants to respond with their digital wallet, which one will they choose?
By the way, I’m betting this is precisely where Personal AI will step in to handle all this stuff for us.
Why? Because the best experience is often no experience.
Maybe Worldcoin is on the right track, but implementation choices are getting in the way
When Sam Altman of OpenAI fame enters the digital ID arena, folks pay attention. And when that arena involves a web3 rando-coin, blockchain and iris scanning, the world’s media - and data protection regulators - wake up.
WorldCoin has been on a rocky journey, mostly getting blocked by various governments. But the project just made some announcements, and there’s a thread to pull on here.
Here are the main themes just talked about by ‘World’ (they’ve dropped the ‘coin’ - I wonder why):
As AI scales, we won’t be able to distinguish people from computers. We’ll soon need the ability to prove we are human. A ‘Proof of Humanity Credential’ (PHC) is going to be critical.
More broadly, we’re going to need ‘Human Verification Services’ (their words).
We need ways to (re)distribute the benefits and gains of AI, as it disrupts and reconfigures our economy (read: job losses and, some say, the need for universal basic income).
I don’t see a problem with that logic.
But many, many people see a problem with scanning an iris to prove you are human, as per World’s choice of techno-optimism.
I’ve heard privately from experts that the iris scanning tech is world class and private. But still. It’s the optics (pun intended, you’re welcome).
Especially when the co-founder is Sam Altman, a man who has been caught lying to the board, and whose other project is, he says, on a course to develop AGI to take over all our human work.
Back to the three points.
I believe these are genuinely the most important, and possibly the most urgent, items to solve across the digital economy. But that the debate is happening in the wrong way.
Is this a question about implementation (how), not the direction (why)?
First, PHC is fast becoming a hot topic. The main questions under debate are
who provides the ‘proof of human’ information
how can we trust the ‘issuers’ of those those proofs (and how does liability work)
how can we do this in an ‘open’ way so that there are no brands, platforms or incumbents able to rent-seek in the middle, nor spy on the data (or access the identity information behind a PHC)
Second, yes we are going to need ‘human verification services’. But again, the question is about how we can do that at scale without a private company or government sitting in the middle (looking at you Google and Meta). It’s why we need protocols not platforms.
Third, on redistributing AI benefits, that’s a harder question (which is a high bar, given the size and impact of the other two questions).
A number of credible people are pointing to ‘tokenisation’ as a way to distribute digital value in a new, and decentralised way. I’m not going to push on that any harder here, but it’s clear (to me at least) that we can’t have #3 without #1 and #2 in place (else how do we know who owns what?).
Which is why I’m bullish on PHCs and digital ID acceptance networks.
It’s why I’m bullish on digital wallets and verifiable credentials.
And why, dear reader, we’re back to Empowerment Tech. Who knew?
Maybe World(coin) is on to something. It’s just that they’ve made some pretty weird-looking (and non-compliant) implementation choices.
What ever happens, you can bet that the world will be watching.
And that’s a wrap. Stay tuned for more Customer Futures soon, both here and over at LinkedIn.
And if you’re not yet signed up, why not subscribe: