The inclusive Machine Economy — Will machines become more human than humans?

In the future, when most functions are taken over by machines, machines will also be tasked with the responsibility to look out for individuals’ well-being in the most moral, inclusive way.

Not all humans are in the same condition and have the same needs, we are diverse, and so, machines must adapt by catering to these differences. Take for example people with physical or mental disabilities, how will machines make their lives easier and balance inequality? Or how about those who cannot express themselves properly due to age, language barriers or because of forced immigration (e.g. refugees)? Will Siri, Alexa, and the other “cool kids” of personal assistants become fault-tolerant to language inaccuracies?

Machines are tasked with making our society more inclusive, lowering the barriers for humans to enjoy an equitable way of living. However, we as humans, are building these machines, and are responsible for designing the right incentive mechanisms to make it economically attractive. For example, making sure a self-driving car will be as likely to transport a wheelchair user from A to B as a healthy well-paid man (who would likely require less effort in boarding the vehicle).

Photo by Franck V. on Unsplash

Similar to humans, machines have an organic operational system (the human brain), cognitive APIs (like the eyes, skin, mouth and ears) and artificial neural networks (instinct and experiences) which allows them to see, recognise and evaluate objects and situations in nearly real-time. Based on the input, a human being can make decisions to help and support other living beings. But what if this task and the huge amount of cognitive impressions that come with it, are taken over by smart agents/machines such as autonomous buildings, drones, caring robots or cars? In that case, we need new incentive schemes for Artificial Intelligence to behave in favour of diversity.

“Only when we succeed in harnessing the technological capabilities of our time to empower every person to become part of our society, we will be able to address the major challenges of the human agenda for the 21st century.” — Marco Richardson, CEO INCLUSIFY AG

This is where inclusive technology comes into play. The society of the future will not be comprised of mostly healthy 20–30 year-olds. I believe the average age of society will increase, and consequently our cognitive + physical capacities, decrease. While today’s inclusiveness is mostly related to race, income and gender; in the near future, it will likely be linked to our physical capabilities, making the range from the standard ‘normal’, to the deviant even more evident.

And lastly, how can we avoid the cycle of history repeating itself and machines following false ethical principles –like the Microsoft Chatbot Tay who made numerous offensive and racist statements?

Machines becoming human companions

But there is hope. With the current advancements in technology, we can assume that (smart) buildings and (autonomous) cars will be equipped with various sensors, embedded algorithms, and software-related functions that can be triggered depending on the physical (or even mental) state of the occupant or visitor aiming to provide a more human-centric experience. Take for instance a user with certain characteristics –like walking impediments– who wants to get from A to B, here, an autonomous car could “read” the signals, and automatically adapt itself using hidden features to become a better machine-to-human companion.

We can go even further and assume that the car (or its digital twin identity) will ask the passenger to grant it access to read and analyse his/her health data (in exchange for tokens) to ensure an even better onboarding experience in the future.

But, why?

The overall question remains. Why should an autonomous car react this way and prefer a wheelchair user over a healthy person where less effort and time is required? What are the benefits? How can we ensure that a digital Hippocratic Oath is operated? We need to re-think what needs to be done so that the Internet of Things becomes the Internet of the right Things.

The incentive machine for an inclusive human-to-machine economy

Let’s try a mind game: imagine if a car would receive cheaper energy during peak zones, or would pay less taxes, or would even get access to software updates much faster than the other cars just for catering to the needs of different humans with inclusivity –would these be good enough incentives for a car to do the “right thing”? Now think deeper, what if the car could become an autonomous organisation itself and would receive a better reputation on the global index of autonomous agents for being the best human companion?

Can blockchain and tokens enable this vision and incentivize machines to do good?

I’m confident that by using blockchain and crypto-economics we can turn this vision into reality.

In his very inspiring speech at the Startup Energy Transition, Trent McConaghy declared the “Incentive Machine” as the most important characteristic of Blockchain and cryptocurrencies.

“You can get people to do stuff with blockchain. How? If they do that thing you want, then you pay them with tokens. Magic Internet Money. “ (10:45)

Let’s assume for one second that the Incentive Machine can also be applied to Machines.

“Macro of motor engine with gears and screws” by Chad Kirchoff on Unsplash

Objectives for the Inclusive Machine Economy:

Alright, that was some sci-fi, mind-blowing writing going on back there. How should this look in practice? I believe we have to design new incentive structures for Smart Autonomous Agents:

1. Let’s incentivise the community to build inclusive technologies. For example, new cognitive API-sets = visualisation of noise, voice control, emotion recognition, speech-to-text.
2. Let’s create new barter-deals among humans and machines. What would machines want in return for inclusive decisions? What are their needs? Cheap energy, compute power, data sets or software updates as units of value.
3. Let’s incentivise digital and financial literacy. Facing the advent of the fourth industrial revolution, we have to educate people about the money system and how they can monetise their own data (for example IoT data) in exchange for cheaper energy.
4. Let’s foster co-existence a.k.a. new income streams. Using blockchain-based token exchanges, a car could constantly contribute to the well-being of its passengers by distributing the economic benefits from its “sensor data trades” or “energy trades” to its humans feet. Sharing is car(ing) –pun intended :-)
5. Let’s democratise and foster innovation. AI experts who were formerly paid by BlackRock or Goldman Sachs to train Algobots for high frequency trading can be incentivised by building Algos who learn how to treat people equitably in exchange for digital or physical commodities (gas/processing power, energy).
6. Let’s be open. The code of any Machine Economy related application should be publicly open and governed by communities rather than by big institutions.
7. Let’s improve the immune system of a decentralised ecosystem. AI experts have to design new reward structures for sanity-keepers which can detect bad Algos from good ones in exchange of other goods and services.
8. Let’s build a digital Hippocratic oath for AI algorithms. Machine learning systems that are trained using legacy datasets could reinforce biases in decision making. There is a document in place, called Ethically Aligned Design, which includes a series of detailed recommendations based on the input of more than “100 “thought leaders” working in academia, science, government and corporate sectors, in the fields of AI, law and ethics, philosophy and policy how to build benevolent and beneficiary autonomous systems.”
9. Let’s use shared, diverse data sets. If our world would ultimately be managed by algorithms (AI DAOs) and populated by self-driving objects, we would need much more data models to reduce the failure rate and the probability of accidents. To increase the limited understanding of the world of an autonomous car and make it less vulnerable to attacks, we should not limit data analytics on the sources of the big players (FAGMA) but should support Open Data Marketplaces where each IoT device or person on the planet has the freedom to trade its data set to AI companies to make algorithms better. Currently Big Data is often biased, because it is mainly based on the input of white males. Hence, it represents only a fraction of our global society. We need new curation markets for Data Models.

“In the same way Bitcoin created an emergent system with the most compute power in the world through open incentives, a properly engineered incentive structure for data would cause the best data in the world for your application to come to you.” as Fred Ehrsam stated in his recent blog article.

That’s why the Ocean Protocol from BigchainDB is such a smart and important building block for a secure and inclusive Machine-Economy.

If we go even further and incorporate the thoughts of the recent visionary paper “Nature 2.0” by Trent McConaghy, we can assume that an autonomous car will ultimately become an organism which lives in an ecosystem of other decentralised autonomous organisations, like self-sustaining buildings or self-owned wind parks.

Take new, emerging protocols +the sheer abundance of sensor, IoT and transactional data + energy and you get the perfect recipe to create new positive dynamics that the ecosystem requires, and thus provide an environment where Humans and Machines can co-exist.

An autonomous car can indeed become more human than the greedy toaster who still has to pay off its gambling debts to the fridge.

Lastly, I wanted to say thank you to my old friend Marco Richardson who inspired me to write this article. We worked together at Microsoft back in 2015 when he was employed as a Microsoft Technical Evangelist for IoT and Hololens on my team, and he has now become CEO of INCLUSIFY AG.

Also kudos to my colleague Carolina Soto (Junior Portfolio Manager of the Machine Economy Investment Cluster) who helped me with sharpening the main messages of the article and Kerri A. Selby for proofreading.

👏🏼 If you enjoyed reading this piece leave us a clap or comment below. We are curious to hear your thoughts!

🤖 We are the machine economy team of the innogy Innovation Hub and believe in a future that is decentralized.

💡 If you are a startup working in the field of blockchain-based middleware solutions, identity or service layer or are just curious about the topic, feel free to contact us! kerstin.eichmann@innogy.com