Putting Fetch.AI’s Open Economic Framework in context

Every individual and business needs a set of rules within which they can operate. This is the case for both laissez-faire markets that have minimal government intervention, and those where governments hold monopolies over entire sectors.

These rules could be about essential functions such as the mode of transactions (be that in fiat or coffee beans) or concern more complex issues such as dispute settlements. The fact that humans’ earliest writings outline debt obligations show how these matters are fundamental to our economies. In the wake of trade wars and markets which are susceptible to changes in human emotion, one wonders if the rationality of a machine-driven economy would see the markets function differently. Financial markets have routinely seen flash crashes owing to bots reacting wildly to specific indicators. However, commodities and stocks have attained relative efficiency in pricing due to ‘intelligent’ agents taking the place of humans trading against each other in stock pits.

Irrational agents from a bygone era

In order for Fetch’s agents to trade value against each other effectively, they need a framework similar to those found in traditional markets. Since these agents will be functioning in mostly digital realms, the rules applicable to them can be designed to be far more effective and ‘smart’ from the beginning. The Open Economic Framework (OEF) is a combination of APIs, directories of services and agents, previous transactions, wallets and agent positions. In summary — it is an essential component of the Fetch ecosystem, allowing agents to search, discover and interact with each other. Much like conventional economies, a digital economy would accrue value over time as the number of transactions, and the complexity of them, increases. The OEF provides an ecosystem for information to be captured within the ecosystem, enabling the ledger to become more useful as the number of transactions on it rises.

Consider for instance, an agent that needs to find information for where data generated from a specific component in cars, in a specific region of the world, can be sourced. Although the discovery process would take computer resources to scan through advertisements and position the Autonomous Economic Agent (AEA) to the node closest to the datasets’ seller, once this is done, the information is relayed across the entire ledger through the OEF. Consequently, the next time there is a requirement for that specific dataset, resource expenditure on the ledger would be drastically lower. This process is comparable to personalised recommendations made from within one’s social network. The theory of Dunbar’s Number suggests that the maximum number of individuals with whom any one person can maintain stable relationships is 150. This indicates the limits on the scale of valued human interactions. By contrast, an autonomous agent is able to tap into the collective knowledge curated through the transactions on Fetch’s ledger and the computing power dedicated to machine learning through mining on the network. This is why the OEF matters; it is a collective amalgamation of the network’s value.

Conventional monetary policy and economic frameworks primarily run on information asymmetry, but interacting autonomous agents could yield better economic output. Today, the rate at which data is produced is unprecedented. The speed at which data is generated means much of it quickly becomes outdated and is forgotten or destroyed before it can be utilised by people who could benefit from it. An autonomous system such as Fetch takes the burden of discovery and trade from the individual and allows it to be automated through agents, while ensuring the privacy of the individual is maintained. Combining financial incentives alongside better forms of value discovery could lead to the birth of entirely new industries and create opportunities for businesses to generate income from sources they have never previously considered.

To take one example, imagine an SME handling a logistics fleet of 1,000 vehicles in India through a Software as a Service (SaaS) offering. Using Fetch’s digital world, the company could realistically generate geographic mapping, logistic time and mileage related data from their vehicles with the integration of cheap sensors. Selling this data to foreign and local companies that are eager to understand the inefficiencies in truck fuel systems, or to those building autonomous vehicles, could prove to be a parallel source of income for the SME. It would also democratise access to the information. Previously, datasets have been accessible only to large businesses or governments. By democratising access, small companies and researchers could discover and pay for data through their agents in order to build unique solutions.

The OEF combines the collective intelligence sourced from the ledger with the automation of data. Much like tracks guided trains in the 19th century and paved ways for the early days of the industrial revolution, the OEF would be the backbone upon which transactions in Fetch’s system scales, collects information and provides guidance. It is not a single rule-set but rather a collection of APIs, on-ledger knowledge (e.g. transactions), wallets and directories interacting with each other. One could draw parallels to traditional economic frameworks such as communism or capitalism, but where human-driven markets are often limited by resources and divergences caused by irrationality, a market-place driven solely by agents could realistically offer an alternative that allows individuals to make better informed choices for themselves.