Enjoy life, study every day and spend time for yourself in Satoshi Club. And today we would like to tell you about the AMA session with our friends from UTU. The AMA took place on September 28 and our guests were Bastian Blankenburg, CTO of UTU, Rahul Srivatsa, Token Sale manager and Chief of Staff at UTU and Jason Eisen, CEO of UTU.
The total reward pool was 500$ in UTU and has been splitted in 3 parts.
In this AMA Recap we will try to summarise the most interesting points for you.
Part 1 — introduction and questions from the Telegram&Bitcointalk community
Mary | Satoshi Club: Hello, Satoshi Club :grinning: Today we are happy to announce that today we have UTU project as our guest :smiley: welcome to Satoshi Club!:+1:
Luis Merino | Satoshi Club: Hello dear community! Welcome to another AMA session with Satoshi Club 🙂
Serg | Satoshi Club: @databu , @Rahul_UTU , welcome to Satoshi Club
Bastian Blankenburg: Hi all, very happy to be here!
Rahul Srivatsa: Thanks for having us!
Mary | Satoshi Club: @databu @Rahul_UTU you are welcome 😀
Irina K. | @satoshi_club: Hi guys, welcome to Satoshi Club
Mary | Satoshi Club: Guys, could you introduce yourselves? And how did you start with UTU project?)
Bastian Blankenburg: I’m Bastian, CTO of UTU. I studied and did a PhD in computer science/distributed AI. This involved multiagent systems using game theory, risk models, trust models etc. It also involved building payment protocols that incentivised agents to adhere to the protocol, before blockchain came along. So nowadays one would use smart contracts for some of this, and I find it exciting which possibilities for distributed systems are now available.
Later worked in industry, then moved to Kenya because of private reasons. There I met Jason Eisen, our co-founder and CEO, who convinced me to join his startup, the taxi app MARAMOJA, because he had this great idea for a trust mechanism. That’s because the taxi sector had traditionally a lot of crime here, and people prefer known drivers. Then from that we spun off UTU, to make its own product of the trust mechanism.
Rahul Srivatsa: I’m Rahul Srivatsa, Token Sale manager and Chief of Staff at UTU. Born in India, but grew up in Tanzania and have spent a good amount of time in Kenya. I have a Masters degree from Boston University in Economics. I has interned with UTU as an economist, working with Bastian on the Tokenomics section of the white paper. Now have transition into my new role. Previously, I have also had experience in the consulting and research world and currently training to be a data scientist. I met Jason a few years ago in Kenya and was immediately convinced about what UTU can bring to the world! Been with them ever since!
Bastian Blankenburg: (Unfortunately Jason has an emergency to deal with right now so he can’t join us.)
Luis Merino | Satoshi Club: Really nice! Thank you for your introductions 🙂
Mary | Satoshi Club: Cool idea and i see that world really needs such service as yours:+1: thank you for intros)
Btw, how is the weather in Kenya?)
Bastian Blankenburg: Right now it’s good, but it’s been raining a lot lately 🙂 The rain seasons are not as reliable anymore as they used to be — climate change.
Mary | Satoshi Club: Yes, we feel the same changes in my country(
Q1 from bitcointalk user apocan99
For providing a “Trustful” ecosystem you “Trust” users, 3rd party people. How do you assure they are trustful enough? What mechanism makes them do the right thing?
Bastian Blankenburg: Actually our mechanism does not aim to tell you who to trust. Rather, it learns which people you trust already for what kind of things, and then provides the right information at the right time to make decisions. E.g. which service prodivers to engage.
But we also have a reward and penalty scheme in our staked endorsement mechanism, which incentivises people to endorse truthfully.
Rahul Srivatsa: It’s about having the right infrastructure to incentivize users to accurate in the information, endorsement or feedback that they provide. We don’t tell people who to trust, but our infrastructure allows users to make more informed decisions on who to trust.
Luis Merino | Satoshi Club: Got it, but how does the penalty system work? Do users face to a great penalisation?
Bastian Blankenburg: Users can stake some amount of UTU Trust Token (a 2nd token different from UTU Coin) on an endorsement. Then, when that leads to a recommendation being shown to someone else, like a friend, and they also endorse, the 1st endorser will be rewarded.
OTOH, if the 2nd user does not endorse but instead disapproves, a small amount of the 1st user’s stake will be burned.
Figuring out the right balance here is the tricky part, these parameters will be subject to a governance mechanism.
Mary | Satoshi Club: But if i don’t have anyone i trust?) For example, i need a nanny, but it’s a new place for me and i don’t have persons i already know enough to trust around?
Luis Merino | Satoshi Club: Exactly! Are users opnions public? I mean, can I see who Mary trusts?
Rahul Srivatsa: Using contextual data available and incorporating it into our trained model, we will show you nanny recommendations from people within your social graphs, contacts, or friends’ friends. Our idea of trust is more human, we trust people like our friends, family or colleagues.
Bastian Blankenburg: Yes there’s potentially a bootstrap problem. But keep in mind that our system is integrated into 3rd-party platforms. So ideally, a nanny app or DApp might already have some data that we could use (and if allowed by users). This has often been the case with our existing clients.
Also, if we don’t yet know any direct relationships of people, we also look at similar groups — “Parents with kids of an age similar to yours in your neighbourhood like this nanny.”
Finally, people are really well connected. At first, we were worried that it takes long to build up social graphs, but it actually goes pretty fast.
Q2 from bitcointalk user juanes
according to new leaked documents, one of your investors, SoftBank, was knowingly letting laundered money to be transferred through its accounts… An investigation is ongoing now. How can this affect your project? What if the laundered money were from crypto?
Bastian Blankenburg: Well, of course having one your big-name investors in bad news is not great. We have to see how and whether it affects reputation. Being German, such things are not exactly news to me because basically all the big German banks have been doing this for a very long time. E.g. Deutsche Bank is probably the biggest and most famous culprit. This may sound like whataboutism, and maybe it is — but I guess I’m really saying that if one needs to raise larger amounts of money, it can always happen that some of the investors turn out to have darker sides.
> What if the
laundered money were from crypto?
If so, it might be another
news item where crypto is associated with bad money and crime. But
again, traditional banks, German and worldwide, are involved in money
laundering much more than probably even exists in crypto. So it’s
all pretty hypocratic I think.
Rahul Srivatsa: Moreover, it is Softbanks subsidiary DeepCore that has invested in us. While this doesn’t diminish the issues around image, it distances us slightly from the issues surrounding SoftBank itself.
Luis Merino | Satoshi Club: Nothing new, traditional finances and economists trying to FUD crypto… 🤦🏻♂
Mary | Satoshi Club: But your relationships didn’t change after all this happened?
Bastian Blankenburg: Not that I know of. I’m not sure how it might affect future rounds. But maybe we don’t need them because we can now continue fully on the decentralised side 🙂
Mary | Satoshi Club: Decentralisation is the main point, i think, it can give us freedom:grinning:
Bastian Blankenburg: Absolutely! We’re very excited to decentralise important parts of our protocol now, and more and more over time!
Q3 from Telegram user @crypto_zones
I see you’ve raised 500k in December last year through bridge found round. Why you’ve decided to raise again and why through crypto? Also, let me know what financial numbers. How do you plan to spend them
Bastian Blankenburg: The news about the 500k round were actually late — we had in fact raised most of that many months earlier. So when the news broke, much of that was already used.
Apart from that, 500k is really not much when you look at what we’re building. There are many parts of the system in active development, and which require research and a lot of innovation.
So our current crypto token sale raises 2.7M if fully bought. Most of this will go into hiring — AI and ML engineers, blockchain engineers, privacy/cryptography expers, general software engineers. But also product people, especially those also trained in psychology, because good UX and UI are just as challenging.
And then there’s some amount set for exchange listings, MMs etc.
Luis Merino | Satoshi Club: How many devs and engineers do you have?
Bastian Blankenburg: There’s 15 people working for UTU right now, plus about 25 working for MARAMOJA, the taxi app from which we originally developed the trust model. Most of the UTU staff are engineers (most full-time employees, some consultants). MM’s engineering team is 6, but we’re also partnering with another taxi tech company. MARAMOJA is still a subsidiary of UTU btw.
Mary | Satoshi Club: Not bad) that’s a big team.
Bastian Blankenburg: Yes, and we’re further growing now.
Mary | Satoshi Club: What will you do with unsold tokens if you will have them?)
Bastian Blankenburg: Those will go to the Ecosystem Growth pool, which can be used by UTU Trust Token (UTT) holders to convert UTT to UTU Coin. To a limited extent, via an auction mechanism.
Q4 from Telegram Username Dixitd00
I really like that besides user recommendations UTU’s trust oracle also accesses off-chain data through social-media channels and uses this data to assess a borrower’s creditworthiness. Could you please elaborate on how this works by providing a concrete example?
Bastian Blankenburg: In general we’re only using data which users allow us to use, because we’re implementing privacy first. So we’ll have to motivate users to provide us data, which is why we’re incentivising user to provide all sorts of data by letting them earn UTT for it.
For lending in particular, there’s the additional incentive to get better terms with the lender through a better creditworthiness assessment, which in turn required more data.
Technically, we building adapters to allow users to provide data from all sorts of places, like for example FB exports. Then whether that data is useful, or likely fake or trash or whatever, is ultimately determined by the data consumers, i.e. our own service but also client platforms like lending platforms. This feedback is again used to determine the UTT rewards for the data provider (user).
Mary | Satoshi Club: So, if i will use your platform you will ask me, for example, connect it to FB? Right? To collect my data
Bastian Blankenburg: Yes, but it’s optional. But you’ll get better rewards if you allow. For FB there’s actually 2 options: provide API access with different data points (e.g. friends list) or upload exported FB data.
Mary | Satoshi Club: It will look like some bounty campaign, user’s data collection?)
Bastian Blankenburg: The bounty is collected by the user, not us 🙂 It’s about giving the user control over which data they allow to be used by which services.
Mary | Satoshi Club: Of course :joy: yes, now i see, how it will work. Thank you)
Bastian Blankenburg: For example, you could say that FB data can only be used by UTU’s recommendation service, but not by the lender. Or the other way around. Or both, or none…
Mary | Satoshi Club: Personally i am very greedy in everything that applies to my data:joy:
Luis Merino | Satoshi Club: Same here :wink::laughing:
Bastian Blankenburg: And you should be 🙂
Mary | Satoshi Club: But with right motivation it will work:+1:
Bastian Blankenburg: Our system let’s you define the terms. Technically, only the access rights are managed on-chain. Data itself can be hosted by any compatible storage service that the user trusts. We’re working with the guys from Dedis/EPFL to adapt their Calypso system for secure and decentralised data access management.
Luis Merino | Satoshi Club: I am that kind of user that reads the privacy policy… 🤣
Q5 from Telegram Username @erol1763
Traditional credit scores are often highly context-sensitive and only apply to specific areas – how do you plan to overcome this limitation with UTU?
Bastian Blankenburg: Our support for lending use cases is highly domain and context-sensitive as well. However, as with our trust mechanism in general, we’re continuously working to identify commonalities across use cases and to make our models more generally applicable. For this, we’re using different kinds of AI techniques, and mix-and-match them to support each client. Eventually, with our learnings and further advances of AI, we hope to arrive at very generalised models.
Rahul Srivatsa: Also, having already piloted with Jamborow, a p2p lending platform based in Tanzania, we are learning more about the various contexts and how lending works in each.
Mary | Satoshi Club: Do you already have any MVP?
Bastian Blankenburg: So far we’ve been onboarding clients on the centralised side one-by-one, and have a few up and running yes. For the decentralised part, we’ll launch the first version of the UTT contract in the next few months (currently in alpha version on æternity testnet).
We’re developing the DeFi-specific oracles and DApps in parallel.
Mary | Satoshi Club: Who can participate in tests?)
Luis Merino | Satoshi Club: Not an exact launch date yet?
Bastian Blankenburg: We’ll soon move to public testing, will announce.
Mary | Satoshi Club: We will follow your news:+1:
Bastian Blankenburg: First thing that we’ll make available is to self-sign up with our API using UTU Coin right after token launch (end of this week). But this first version will not have on-chain endorsements yet. (and therefore the UTT incentives)
Mary | Satoshi Club: Good luck with this, sure that your team will make its best to deliver the best)
Q6 from Telegram Username @Brainchest
How does the mechanism of trust distinguish an artificially created human reputation from a real person, what is the result based on and what sources are used for it. Are you responsible if the system is wrong and the platform who trust your service loses money? For example, if I want to take out a loan and specially take it several times in advance, then return the loan, which will give me a good reputation (credit rating), in social networks I will create a profile of a successful manager with text keys about a large salary, which will give the robot false knowledge. As a result, with an excellent reputation, I will ask for a loan and will not give it back.
Bastian Blankenburg: Wow, all the hard questions that we thought hard about wrapped in just one :laughing:
> How does the mechanism of trust distinguish an artificially created human reputation from a real person, what is the result based on and what sources are used for it.
We’re tying feedback to specific service transactions. So if you want to endorse a service provider, the system needs to have registered a previous transaction where you actually booked that provider. There will be some exeptions, like a mechanism to upload legacy data, but that will require some level of KYC then.
> Are you responsible if the system is wrong and the platform who trust your service loses money?
Yes and no. We’re actually in the process with existing clients to devise success-based fees for our services. We would not provide collateral ourselves however. I.e. in the worst case, our fee might be zero.
However success-based pricing is often difficult and depends a lot on the concrete use case, and we also need to protect ourselves from fraud.
> For example, if
I want to take out a loan and specially take it several times in
advance, then return the loan, which will give me a good reputation
(credit rating), in social networks I will create a profile of a
successful manager with text keys about a large salary, which will
give the robot false knowledge. As a result, with an excellent
reputation, I will ask for a loan and will not give it back.
Yes
that’s a classic problem in the lending space. Which is why past
loans are never the only data to look at when determining
creditworthiness. But as mentioned in the previous question, support
for lending platforms is often very context-sensitive. There’s too
much cases and details to elaborate here, please talk to us directly
if interested 🙂
Luis Merino | Satoshi Club: Is KYC compulsory to be able to use the platform?
Bastian Blankenburg: Not for basic usage, but for some functionality yes.
Part 2 — live questions from the Telegram community
Q1 from Telegram user @starshades007
UTU focuses on TRUST, but to be effective, there must be a TRUST RATING INDEX so users don’t get astray, do you have this
Bastian Blankenburg: We believe that trust indexes are not helpful. What does it mean if we told you that you can trust this car mechanic to a degree of 7.68? Will he repair your car to a degree of 7.68? And what does that mean?
Instead, we think it’s much more useful to show “Your friend Henry, who has the same car, really likes the quality of this car mechanic.”
Q2 from Telegram user @vnmb788
You organized an AMA session very rewarding and received a lot of questions related to utilities and technology, future vision, …
So now I want to ask what do you want to receive from the community? The crypto market has a lot of fiercely competitive. What strategies do you have to attract users, and keep them stay with your platform for longterm?
Jason Eisen: We want the community to be as inspired to improve digital trust as we are. This is our life’s work and the implications of the system we build are too great to do it alone, too important to get wrong. We are looking for the community to engage and vote on key issues, to help us build, grow, and deliver on our vision to become the trust infrastructure of the entire internet.
Q3 from Telegram user @AugusS7
What is really the mission and vision of UTU? Is it simply to develop an infrastructure that provides the best recommendations and services to exchange platforms? and with her achieve the adoption?
Bastian Blankenburg: Not only to exchange platforms, to all sorts of online marketplaces. We want to become the default trust layer to integrate into any marketplace app. Like how many online platforms integrate a Paypal button now for payments, we aim to make them integrate UTU’s recommendation widgets a step earlier, when the user makes their choice.
Q4 from Telegram user @huydo
Why are there dual token mechanism between UTT and UTU, please kindly explain roles of each in ecosystem?
Jason Eisen: the idea behind the dual token system was to separate money from trust. That people should not be able to buy trust but should be able to earn from building trust.you can read more in our white paper: https://docs.google.com/document/d/e/2PACX-1vSn7d9BovABf8IxaK9uvvoUkRQKMMEsUqxvetEjovhq-F2Sw7czzbBV4Gghsoi1txn6ydSx3bqeAXst/pub
Q5 from Telegram user @Aleurich
Why do you consider that rating the platforms according to their level of trust is important? Isn’t it more convenient for each person to do their own research?
Bastian Blankenburg: Yes, that’s why our mechanism learns who trusts whom for what kind of recommendation in which context. Most people, when having difficult decisions to make, call someone that they trust in the matter and ask them for a recommendation. “Hey, do you know someone who’s good at …?”. Our mechanism is built to provide this information very conveniently, at the right time and place.
Q6 from Telegram user @BboyForteVZLA
Can you tell me how to participate in your token sale? What requirements do I need? Is it available to all or is it restricted to some?
Rahul Srivatsa: I’ll take an easy one:
1) Please join out TG channel: https://t.me/UTUtrust
2) Find the pinned message describing the steps required
3) Complete the automated KYC and wait to get approved
4) SAFT will be sent once you have been approved and then you have to complete it within 24 hours
5) Then you sit back and relax until the TGE!
In terms of restrictions, we do have geographic restrictions only due to legal issues surrounding crypto in these countries.
Q7 from Telegram user @RoZerius
How difficult is it to fullfil legal requirements from governments to be able to trade with fiat currencies? Are there countries where UTU cannot be used yet due to those regulations?
Jason Eisen: Staying legal and compliant does indeed require much deliberate effort and resources (would be great if someone could make it easer) but it’s just part of the cost of doing business. We are running for years and differrent countries and always have to be aware of the regulatory environnment and how it affects your business. For us, as a trust company, we think it’s pretty important to stay compliant in everything we do and you’ll often see us make some tough choices (that perhaps even get community blowback) in pursuit of compliance, fairness, transparency, and trust.
Q8 from Telegram user @KeymerS
To be able to enjoy UTU services, do they have to be totally Defis projects? Or can totally centralized projects get involved in it and through UTU achieve decentralization?
Jason Eisen: we offer our services across sectors and across both the centralized and decentralized worlds. Our idea is to bring UTU to where the people are and help be the bridge that transitions them from the centralized to the decentralized.
Q9 from Telegram user @Jonahapagu
One of UTU’s Notable partnership is with Graychain(now called credmark) to deliver trust-powered, privacy-preserving, borderless Crypto Credit across africa and enhance credit scoring on the blockchain.. Can you tell us how UTU’s Trust Ecosystem is Reliant on this partnership specifically with respect to credit scoring
Jason Eisen: UTU and Credmark work closely together to combine our data sets and analytical tools to augment each others’ analyses and services. Look for more news about developments from this partnership soon
Q10 from Telegram user @Gutike95
In the same way that the UTU Trust Portal provides recommendations on third-party oracles, dApp and Exchange, it shows recommendations of which ones not to use and which ones are risky or scams?
Bastian Blankenburg: Our mechanism includes disapprovals, which will be used to show warnings, yes. However the threshold for this is higher than for recommendations, to prevent too-easy sabotage. Please read the Penalty section in our whitepaper for details. https://docs.google.com/document/d/e/2PACX-1vSn7d9BovABf8IxaK9uvvoUkRQKMMEsUqxvetEjovhq-F2Sw7czzbBV4Gghsoi1txn6ydSx3bqeAXst/pub
Part 3 – Quiz Results
In the final part we tested your knowledge in terms of UTU. They’ve prepared 4 questions for this part. The total reward pool for quiz was 300$.
For more information and future AMAs, join our Social Media channels:
English Telegram group: https://t.me/Satoshi_club
Russian Telegram group: https://t.me/satoshi_club_ru
Spanish Telegram group: https://t.me/satoshi_club_spanish
Telegram Channel: https://t.me/satoshi_club_channel
Twitter: https://twitter.com/realsatoshiclub
Website: https://esatoshi.club/
Our partners:
UTU Community: https://t.me/UTUtrust