Published in


Fetch.ai Community AMA Recap (8th December 2020)

Thanks everyone for joining for our Community AMA. In attendance we have

Humayun Sheikh — Founder+CEO

Toby Simpson — Founder+COO

Maria Minaricova — Director of Business Development

Jonathan Ward — Chief Technology Officer

We will start in a couple of moments

What is in store for today’s AMA? We’re going to do a short update on some stuff we’ve been working on — notably the Collective Learning on COVID-19, Field Trials on Mobility in Cambridge and Munich, and finally on the roadmap for 2021.

Following these we’ll unmute the channel for questions, and we’ll do a couple of rounds of those, so that we can cover as much as we can.

Please note this AMA is all about Collective Learning, Fetch.ai’s ongoing mobility trials, and development for next year. We cannot discuss any price activity, but you already know that 👍

Toby, Humayun, Maria, Jon -


So first off — let’s say hi to our community! Then we follow up on what is Fetch working on at the moment — especially with respect to Collective Learning and Mobility?

The floor is all yours.

Hi Everyone, glad to be here with our community! (Humayun)

Hello everyone! Hope you are having a fantastic day! Great to be here again with you guys!! (Maria)

Hi everyone, welcome to today’s AMA and delighted to be here. It has been an amazing couple of weeks for agents: our incentivised testnet delivered well over a thousand agents online at once, all over the world, and we’ve been out and about with our DDN for the first time — driving about and seeing all the pieces of the puzzle function together. And there’s so much more to come! (Toby)

Hi everyone — looking foward to hearing your questions! (Jon)

A: The Early Access Program has recently launched. We will be selecting a few waves of users to join. Initially we were focused on professional / experienced traders in the crypto/commodities space, but now we encourage anyone interested to register. The first wave was small — 5–10 people — but we have now rolled out to over 50 people and will be rolling out even more over the rest of November and December. This process is allowing you to pioneer the platform and provide us with the crucial feedback to get it as ready as possible ahead of the full launch. We are launching a trading competition very soon after which we will be opening the liquidity pool in January and ultimately the launch by end of January audit and regulation permitting. (Humayun)

A: If there’s one thing that anyone who works in technology knows, its that you never quite end up where you think you would, you end up somewhere slightly different, but that’s not a bad thing, that’s adapting. Largely, with what we’ve done with the underlying network, the OEF, the agent framework and collective learning has exceeded our expectations. The agent framework in particular is an incredibly powerful toolset, and is about to take another leap in what it can do, how easy it can be used and more. It’s very exciting: it’s the new decentralised, autonomous economy, and we’re building and deploying it. (Toby)

Great question! This is is a very cutting edge project in a cutting-edge space so our goals have shifted a bit over time. We’re making great progess with the core objectives in multi-agent systems and machine learning — which are both hugely powerful and are going have a huge impact in the near future if we keep up the rapid progress. (Jon)

A: It is planned for early January 21. (Humayun)

A: Thanks! It’s great fun to use and very easy, either from a service provider’s perspective or from a user’s perspective.

Different service providers have different regulatory and legal frameworks, for example; if you’re driving people around, depending on where you live depends on how you must be licensed for that to be possible. If you’re moving packages around, it’s more of an insurance issue than anything else — pretty much anyone can pick it up, use it, and deliver value at their convenience. (Toby)

A: Been there, did that 🙂 And the results of it are going to be shown in the coming week or two. We popped the app onto a couple of phones, and we delivered ourselves around Cambridge, whilst the agents did their magic in the background. With all the work that goes into creating all these moving parts, it’s kinda amazing to see it all do its stuff for real. (Toby)

A: Yes, we are setting it up right now. Will be releasing more info soon. (Maria)

A: This is aimed for December and will be announced very soon. All FET holders who participate in FET staking don’t need to worry as they will be rewarded automatically. (Humayun)

A: Yup, we have a section of incentivised testnet about to launch on this very subject that @Jonathan6620 can talk a little more about. Again, a very important step towards true decentralisation where the users of the network are its owners and controllers. (Toby)

A: We’re just putting the finishing touches on the documentation and software for the governance test-nets. These will be released in the next few days. The goal is to introduce the community to some governance questions so that they can start getting involved in how the project develops over time. (Jon)

A: You can see the Fetch address format on the block explorer for out test-nets: https://explore-agent-land.fetch.ai/transactions. …and also by participating in any of our incentivised test-nets: https://docs.fetch.ai/i_nets/. In the next round, we’ll have some challanges for non-technical people to get to grips with our wallets and block explorers. (Jon)

A: … and we have a bunch of code and tools that allow you to create these things. If you’re a developer, this is super easy to do in Python or other languages. (Toby)

A: We are working with AI data scientists and research groups in the space to enable these tools to larger audience, as we believe they would be hugely beneficial to many many healthcare projects and organizations. (Maria)

A: thank you @riseofphoenix ! as always we rely on your support to improve awareness of our projects. We are engaging with several healthcare participants to deploy a commercial solution. (Humayun)

A: That’s a great question. There are a few ways of doing this; it’s possible they can have a public identity with something like a SLA, we’re also looking into reputation and dispute resolution systems so that agents can have confidence in the data they’re buying. (Jon)

A: Very good question.

And trust is delivered in layers. You’ve got your history of being trustworthy, you’ve got the potential for stakers to be involved to provide incentives to be honest, and you’re potentially compared to those around you doing similar operations. Then there’s digital identity: proving who you are and associating that identity with what you provide. This, along with verifiable credentials to prove you’re entitled, qualified or able to provide the information can make a significant difference. (Toby)

A: We continuously working on this in the background. Now with the release of Collective learning tools it will be even more interesting to engage with new partners and show them these fantastic tools and bring them on board! Absolutely! (Maria)

A: I come from a computer games background where I wanted to build bigger and bigger digital worlds, and was hampered by technology. Even with an agent-based approach, and metaphors from biology, a client -> server trust model prevented infinite scale. Sharding and other technologies helped, but you ran out of steam and were never really using all the potential computing power provided by the users of the network. Blockchain and other cryptographic technologies solve this: it means that we can build massive worlds, we can populate them with endless populations of agents, and we can make better use of the assets that we have by connecting them actively to each other using AI and ML. Quite honestly, this underlying infrastructure and capability is about as close to the worlds I wanted to build as is possible. (Toby)

A: Proceeding as part of the larger mobility picture. Signs, as an augmented reality for self-driving cars and traffic optimisation system, coupled with the parking and zoning work, the DDN and more all lead to being more effective with mobility assets. The potential to optimise congestion out of the network, whilst delivering an incentive model for everyone to be able to take part is one of the exciting parts of it: it’s not just one or two large commercial entities involved, it’s anyone who can get involved and benefit from doing so in the form of passive income at the very least, or more for those wanting to engage. (Toby)

A: We continue the work on this prototype, and there is currently more development being done that we should be able to share with you in Q1 2021. (Maria)

A: For the foreseeable future, yes, but the returns will eventually be earned for securing the main-net which is due to be released next year. (Jon)

A: Our plans are to deliver technology that can be use all around the world, including the US.

In Europe we are engaging with our regulators and policy makers through Blockchain for Europe association. In other parts of the world we are following closely and evaluating what is going on and how are the regulations being set up, used and what is the response from these innovative AI and blockchain solutions. It is not always straight forward, there is more work that needs to be done. But more and more people realize that decentralization has some key benefits — such as data privacy, enabling AI for all and a more democtratic way, and so we see support in our actions and willingness to listen and learn more about the technology and benefits. (Maria)

A: We always try to keep up with the regulation. As for US we have done a few campaigns to bring awareness to Fetch over the last couple of months. Our Covid trial also had deployments in LA 😉. On the tech and market size US is a very important market and we are looking to improve our reach in US. (Humayun)

A: It’s due to be launched in March 2021. (Jon)

A: We can’t comment on individual companies but the collective learning stuff is particularly well suited to big pharma. (Jon)

A: We’re working to fix that, btw. Particularly with agents: we want deploying and working with agents to be trivially easy. You want to manage a resource you have? Make it available to others? Well, when we’re done that’ll be clickity-click-clickity-click-**deployed**. (Toby)

A: Running a validator is quite a skilled job — you need to be able to run nodes reliably for long periods of time and also generally “take care” of the network. We’re planning on running a test-net early next year so that anyone who is interested can give it a try. (Jon)

A: Sounds good to me 🙂 I have a screenshot somewhere showing a visual development tool that could make agent development “Scratch” like: drag, drop, connect. With higher level components, that gets easier and easier. With agent hosting taken care of, there really is nothing to worry about beyond identifying a need. (Toby)

Alright, thanks very much Humayun, Toby, Maria, Jon.

We hope the community enjoyed this AMA. Stay tuned for more updates.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store