By Lubna Yusuf1
1Founder, La Legal
We are surrounded by warnings of a media dystopia, led by artificial intelligence, which will disrupt the way we perceive the world and ourselves. The boundaries between real and virtual realities are fast diminishing and the sci-fi movie scenarios of our youth no longer seem so far-fetched. With algorithms making decisions for us and data systems constantly monitoring us, where will machine learning and deep tech lead human cognition? Are we destined to lose our learning capabilities, with machines programmed to do all the “thinking”? If future decision-making is based more upon results of data analytics than on human reasoning, a whole new concept of human intelligence is set to evolve. The future that awaits us is the human-AI convergence: a whole new interdependence between ourselves, machines, data and numbers.
In recent times, Google pulled out of a Department of Defense contract and called on the US government to formalize rules on the use of AI and Amazon joined Microsoft in calling for regulations on facial recognition. While seemingly significant now, corporate stances of this kind may soon be nothing more than a relic of the past, if privacy becomes a lost concept as the commercialization and trading of data become more commonplace.
In much the same way as blood and sperm banks exist today, it is likely that in the future there will be data banks stocked full with data (such as biomedical data and human behaviour patterns), to fuel sophisticated analytics in a vast array of fields. People will be able to knowingly and freely deposit their human data in these banks for research and deep learning performed by machines. It will also be possible to sell and earn revenue from personal data transactions. However, it will be interesting to observe how the data collected will be segregated and classified according to the race, region, economic and educational background of data sellers and donors, and the uses to which the data will be put.
What privacy laws shall prevail? What will be the quantum of ethics used to demarcate the rules of engagement with these data banks? Such are the questions raised by the use of data and AI in the new internet world.
AI lacks all five human senses of sight, smell, touch, hearing and taste. Nonetheless, today AI already decides which advertisements appear in your newsfeed, which song to play next, which direction to take and your next meal. Human intelligence and our sense of compassion separates us from other species. Our sense of smell is the strongest link to our memories. However, our future memories will be strongly manipulated by AI-driven systems being fed with data to increase consumption by consumers. AI may be disrupting the world around us, but some things never change: the underlying commercial objective is always and forever to sell more of everything and anything. The fine line between popular influence versus AI-generated influence will be diminished and our individual decision-making will be largely impacted by the data that will have been fed into the system.
In a world of growing social media influencers and pop “Insta-celebs”, the most influential “influencer” will be the analytics-driven AI itself, which will decide for us before we can decide for ourselves. Online surveys interpret our motivations and desires to take us to advertisements for our next holiday, and that sweater we browsed on Amazon keeps flashing on every tab we open, even when we browse from a different device. Worryingly, almost 90% of apps and extensions lack an approved privacy policy, signalling that the people behind the creation of these apps and extensions either don’t understand or don’t care about the concept of privacy, and there is a lacuna of internationally harmonized regulations and laws to define data privacy norms.
In this context, who decides “how much AI is too much AI” and when is that likely to happen?
A team from Columbia University has already developed a system that combines deep learning with a speech synthesizer that translates one’s thoughts to words. As our cognitive capabilities are increasingly heavily influenced by data-driven AI, our perception of the world around us will also be impacted. With the huge influx of apps, bots and ever-increasing connections across the world, we are already witnessing a rise in the number of people feeling “disconnected”. However, perhaps this phenomenon is not so surprising if we place it in the wider context of deteriorating mental health and depression. Instead of it causing further deterioration on our social fabric, could AI possibly be a step towards the creation of a technology to decode people’s thoughts for their own benefit? There is already a huge change in the order of the realities as we know them, with major changes in the interdependency of the virtual world on our real-world happiness and emotional quotient.
We are “happy” if we get liked or loved on social media; we create virtual worlds for ourselves and our well-being is influenced all day every day by strangers, as we share feelings in this virtual space through emoticons and emojis, rather than through actually smiling or speaking in the real world. People are increasingly dependent on such virtual spaces for personal validation, relationships and even sex. This has significantly decreased our faculty for language and facial expression as a means of communication. The human mind has learnt a whole new language of the internet, with symbols replacing words as the primary means of expression, communication and manifested emotion. This is both fascinating and alarming because we are possibly the last generation to experience both the internet and non-internet worlds of communication.
Will future generations come to use their opposable thumbs to only type on mobiles? Will writing and drawing on paper become entirely obsolete?
Humans have memories and up until this century, all of mankind’s history had been preserved through cave writings, books and audiovisual formats. However, with the advent of the internet we have a new medium of data transfer and storage that makes data destruction exceptionally complex and its multiplication potentially infinite. APIs are currently connecting anything to everything. In the future, this may have irreversible consequences. Wealth and power will be wielded through the ownership of and access to vast quantities of data. In this context, how safe is anyone’s data and who bears the risks and responsibilities for its use and protection? Beyond the GDPR, we can expect a further raft of data-related laws to emerge. In the future, data theft may actually be the cause of a major international conflict, with a “data war” being used to destroy economies through the mass manipulation of advanced biometrics, bank details and the sensitive personal data of a nation’s population.
To live consciously, think rationally and interact emotionally is what makes us human; and we must ensure that, as humans and in spite of any AI-induced interference, we retain control of our cognition. Anything less, and we may end up being the very last of the Homo sapiens, particularly if we stop learning and evolving, and delegate our agency by default to machines for them to do the learning and thinking for us.