AI is poised to transformed business in the most revolutionary way since the impact of computer technology in the late 20th century….I know what you are thinking, We’ve seen this before: E-commerce comes to mind. We couldn’t pay for products over the internet until the early 1990’s…But E-commerce was an evolution not necessarily a revolution. For sake of context (let’s defined the two). Evolution refers to the gradual development or changes in something over a period – while Revolution means ‘a turnaround’— a sudden, complete, or radical change in something. In 1982,  Boston Computer Exchange became one of the first E-commerce platform when it served people who wanted to sell their used computers.   So it’s debatable when E-commerce started – some will argue it started way back “When”.   

Through its algorithmic behavior, Artificial Intelligence (AI) powered by machine learning and natural language processing has the ability to identify patterns in data to accomplish feats that was once considered impossible. At least impossible within the boundary of time that such feat is considered useful.  Of course, if you wanted to wait 50 years to determine which corporations that evaded their taxes last year, then prior technologies would suffice – but time is “money” so they say.   Or if you were trying to predict the spread of an infectious disease – AI can certainly do that 100 times faster than the most skilled epidemiologist.  Researchers predict that AI will double annual economic rates by 2035 by changing not only the type of work but how we do the rudimentary tasks of today.  How a user ask for help and to whom the question is asked will fundamentally be changed through AI algorithms.  Spawning a new set of relationships between man and machine.  

The Scary

Some would consider the risks and perils of AI – as even more dangerous than the radical value of the technology at its best.  New Zealand made history earlier this year as the world’s first country to set standards for government use of algorithms.  The European Union (EU) is proposing one of the first laws globally to regulate the use of artificial intelligence for applications like hiring and policing.  Now I for one being a Black Man in America applaud that – the American criminal justice system has for  over 400 years subscribed to the  premise that “All Black Men Look Alike”. Law enforcement services are cleverly and secretly employing software to track people’s faces, over saturate police officers in areas where crimes are most likely to happen, not-so-randomly stop make and models of cars that certain ethnics groups are most likely to drive – all with sophisticated AI algorithms.  I know what you are saying, “I’m rarely targeted by the police”…Well what about this one – Utilities companies “can potentially” leverage voice analytic software to determine your ethnicity and then adjust your rates based on where you live and who you are.  We know insurance companies for ever have based their actuarial principles on ‘who you are’, ‘where you live’, ‘what type of car you drive’, “how likely you are to speed” and ‘how likely you are to die’ – now imagine an AI algorithm doing it for them – with little or no human intervention or oversight.

The Real

Racial categories are, by definition, unequal categories. They reflect not universal truths but historical processes that have linked racial status to economic, political, and social inequalities. So if artificial intelligence practitioners train and deploy models using these artificial categories, the categories will be reified and inequality will be reproduced.

But even if  AI practitioners do not pay attention to racial categories, they will still reproduce socially embedded inequality by default.  So we must design a more equal and just AI, one that recognizes people’s chosen identities while avoiding bias and discriminatory practices.  In order to do this we need structure, guidance and practice and in some cases we need “oversight”.. Something that will help us be our best selves-

There is a saying in computer science, something close to an informal law: garbage in, garbage out.  It means that programs are not magic – especially at the information layer –  If you give algorithms flawed information, they won’t fix the flaws, they just process the information. “It’s discrimination in – discrimination out.”

I know what you are thinking “It’s The Scariness of Realness”.