We are walking towards an artificial world, a place that raises questions about what’s like to be a human being. Technology and human race has never reached such symbiotic relationship; and it has only started.
A CONTINUOUS (R)EVOLUTION
The ability to automate processes is a uniquely human characteristic. With the help of artificial intelligence, we are at the beginning of an unprecedented revolution in terms of scope and magnitude, one with the potential to transform the economy, mobility, politics, healthcare, security and society in general. Will machines take our jobs? Yes. But for the most part, the thievery will be limited to the boring, mundane and repetitive tasks, leaving us with time and energy to focus on what we actually enjoy doing. Although it comes naturally for people to recognize their friends, it can be tricky for machines; straightforward tasks for machines like analyzing gigabytes of data, remains impossible for humans. By automating jobs that involve exhaustive repetitive human labor, society will shift towards jobs that require soft skills such as communication, collaboration, and creative problem-solving, fundamentally changing the dynamics of production and work, potentially reshaping the very meaning of life as we know it.
As a possible new infrastructure for nearly everything, deep learning is already present in many aspects of our lives, yet we are largely unaware of its existence and impact. With the exponential growth of data everywhere, humanity depends on intelligent machines to extract, digest and present this data in meaningful ways. From constant gradual improvements in machine translation services, voice-activated assistants, and personalized recommendation engines, to personalized ads and algorithmically generated news stories, machine learning is embedding itself in our daily lives with or without our input.
These changes are so broad and fundamental that it is difficult to find comparable narratives in human history. Like industrialization, machine learning will probably change society in ways humanity will only be able to comprehend after the (r)evolution has already taken place. Nevertheless, it is possible to ascertain certain overarching trends that simultaneously pervade a number of areas and/or industries.
A Matchless Milieu
Before industrialization and large-scale automation, most interactions - private, business-related, or otherwise - were highly personalized. The local bank managers knew their customers on a first name basis and the tailors and shoemakers made their products specifically for a specific individual client. Industrialization and mass-marketing began to erode the personal connection between customers and industry, grouping consumers into large flocks, targeting their marketing to the least common denominator of the whole group. This is one of the explanations behind the need to dumb down media, TV, and radio broadcasts in order to appeal to the masses.
We are now entering what is being dubbed the third great wave; personalization is making a comeback. Products are shifting from being mass-marketed to hyper-customized using machine learning and data-driven personalization. With self-driving vehicles and shared autonomous air taxis, entire communities are adapting to efficiently serve the needs of each individual, while saving energy and money in the process. Smart cities could be equipped with urban sensor networks that connect public and private services to one another, rewarding citizens for good behavior, and ranking them with personalized scores. Citizens with higher scores would be able to access special reserved public services. Inhabitants could be rewarded with a digital currency for every time they collaborate to maintain and improve the city.
Cities resemble an organism, a decentralized network of nodes that are highly interconnected, similar to the composition of the brain. These nodes could vary from human consciousnesses to AI, to sensors that relay information from the physical world to the network and actuators that translate and transform signals from the network into physical changes. By cross-checking personal and big data and running simulations, predictive analytics could allow us to be better prepared for natural or human-made disasters, allowing us to predict and prevent instead of simply reacting.
UNDERSTAND AND PREDICT
An Embedded Omen
A famous deterministic thought experiment known as Laplace's Demon suggests that if we were aware of the precise location of every atom in the universe at any given moment, we could apply the laws of physics in order to determine their location at any other specific time, thus predicting the future. Deeper understanding of available information stemming from the capabilities of machine learning algorithms could be moving us in exactly that direction.
Artificial agents are constantly monitoring our bodies, our behavior, and the environment, while classifying data and optimizing processes. Specialized in finding patterns, these algorithms are already developing intricate solutions to problems that are beyond human understanding. Imagine cross-referencing genetic data with wellness habits, food intake, and city pollution levels, to determine susceptibility to certain conditions and potentially learn how to mitigate the issue before it escalates. If done right, fluid communication and data-mining entire communities in real-time could create a more inclusive and diverse community. If improperly deployed, we could increase socio-economic gaps and split society into a small super-privileged part and a large low-income class.
There is reason for hope, worldwide middle class growth is changing the historical income distribution visualization, becoming less like a camel with two humps and instead becoming more evenly distributed.
The Birth of Ersatz
Automatic real-time translation could soon mean a world without language barriers. If we can represent concepts numerically then maybe language becomes an implementation detail and we could choose to "render" a certain idea or concept to a language and writing style of our choice. Brain-machine interfaces could allow us to bypass keyboards, touchpads and even words altogether, letting us communicate through pure thought with each other and potentially even with other species and artificial agents. Ideas, texts, and notions could be communicated via semantic representations.
Brain implants could require learning a new modality of perception. Because the brain is still forming its connections, new senses and powers could be incorporated from an early age, and perceived in the same way as natural senses. Younger generations may learn to sidestep natural language altogether. Memory and thoughts augmented to the extent of the whole network. Looking up information on the internet could feel as natural as the sensation we get when touching something with our fingers. As soon as a seamless connection between the brain and the network is established, the very idea of physical location will lose significance. The sense of individuality dissolves.
However, people who have access to or want to use sensory augmentation would hold a tremendous advantage, nearly being considered as a different species, or an evolution of humankind. Inequality among people with different augmentation levels might grow, making it difficult for different groups to even communicate with each other. Machines are becoming so complex that they might not serve exclusively as tools, but become agents themselves, collaborating with humans in creative, undiscovered ways. Centaurs, or human-AI teams, are better than humans on their own or artificial competitors by themselves. Working in tandem towards the greater good allows both parties to supplant deficiencies from their partner.
At the moment, all artificially intelligent systems are "narrow AI" meaning that they are specialized in restricted fields, but given enough complexity, sophisticated AI could hypothetically outperform a human in any possible field, evolving and perfecting its own architecture. This agent is being called artificial superintelligence and it would likely be humankind's very last invention, as by definition, it would be better at inventing things than any human. It is of the utmost importance to invest in research to ensure that such a powerful machine would share our moral values. Human-machine interaction ethics could be codified in the form of smart contracts. The Quran and other sacred scriptures could serve as intelligent blockchain auditors for any kind of transactions and exchange. Religion, instead of losing significance, could become the basis for codifying moral rules and ethical standards for humans and machines alike.
However, we are imperfect beings subject to biases. If we program our flaws into a super intelligent self-improving system, these flaws could escalate and flourish in unpredictable manners. As Stephen Hawking said, "Unless we learn how to prepare for, and avoid, the potential risks, [superintelligent AI] could be the worst event in the history of our civilization." Yet, if well implemented, it could help us come up with solutions to humanity’s biggest problems - specifically climate change, world hunger, conflict resolution, and religious and individual freedoms. The key is to reinforce alliance and cooperation.
A confession: human values will need to be emulated in every relationship between humans and machines, creating a powerful affiliation. Society will have to make a concerted effort to collaboratively work alongside machines, without either side attempting to replace the other. Open acceptance of this partnership will lay the foundation for our future existence and even the continuation of life itself.
By using Envisioning’s proprietary methodology Readiness as a means to generate a quantifiable indicator of technological progress, the following research project is a continuous collaboration between the Technology Foresight research program of the World Government Summit and the research institute Envisioning. Each of the 109 mapped technologies were ranked and plotted on an imaginary timeline that classifies them as an idea, a concept, a prototype or a product, helping navigate through possibilities in the near and long-term future, through the year 2071. These technologies were organized in affinity groups according to their main industry focus, and thus corroborate to design future scenarios that we have envisioned. Check all of our curated clusters with mapped technologies, and discover the main leads for the future of Artificial Intelligence.
Envisioning is a virtual research institute. We provide technological foresight to policy and decision makers worldwide. Our global team of academics, hackers and designers study technology to understand accelerating change. We share our work with our own tools, methodology and design.