Importance of AI and data literacy

AI and data literacy refers to the holistic understanding of the data, analytic, and behavioral concepts that influence how we consume, process, and act based on how data and analytical assessments are presented to us.

Nothing seems to fuel the threats to humanity more than AI. There is a significant concern about what we already know about the challenges and risks associated with AI models. But there is an even bigger fear of the unknown unknowns and the potentially devasting unintended consequences of improperly defined and managed AI.

Hollywood loves to stoke our fears with stories of AI running amok over humanity (remember “Hello, Dave. You’re looking well today.” from the movie 2001: A Space Odyssey?), fears that AI will evolve to become more powerful and more intelligent than humans, and humankind’s dominance on this planet will cease.

Here’s a fun smorgasbord of my favorite AI-run-amok movies that all portray a chilling view of our future with AI (and consistent with the concerns of Henry Kissinger and Stephen Hawking):

• Eagle Eye: An AI super brain (ARIIA) uses big data and IoT to nefariously influence humans’ decisions and actions.
• I, Robot: Cool-looking autonomous robots continuously learn and evolve, empowered by a cloud-based AI overlord (VIKI).
• The Terminator: An autonomous human-killing machine stays true to its AI utility function in seeking out and killing a specific human target, no matter the unintended consequences.
• Colossus: The Forbin Project: An American AI supercomputer learns to collaborate with a Russian AI supercomputer to protect humans from killing themselves, much to the chagrin of humans who seem to be intent on killing themselves.
• War Games: The WOPR (War Operation Plan Response) AI system learns through game playing that the only smart nuclear war strategy is “not to play” (and that playing Tic-Tac-Toe is a damn boring game).
• 2001: A Space Odyssey: The AI-powered HAL supercomputer optimizes its AI utility function to accomplish its prime directive, again, no matter the unintended consequences.

Yes, AI is a powerful tool, just like a hammer, saw, or backhoe (I guess Hollywood hasn’t found a market for movies about evil backhoes running amok over the world). It is a tool that can be used for either good or evil. However, it is totally under our control whether we let AI run amok and fulfill Stephen Hawking’s concern and wipe out humanity (think of The Terminator) or we learn to master AI and turn it into a valuable companion that can guide us in making informed decisions in an imperfect world (think of Yoda).

Maybe the biggest AI challenge is the unknown unknowns, those consequences or actions that we don’t even think to consider when contemplating the potential unintended consequences of a poorly constructed, or intentionally nefarious, AI model. How do we avoid the potentially disastrous unintended consequences of the careless application of AI and poorly constructed laws and regulations associated with it? How do we ensure that AI isn’t just for the big shots but is a tool that is accessible and beneficial to all humankind? How do we make sure that AI and the massive growth of big data are used to proactively do good (which is different from the passive do no harm)?

Well, that’s on us. And that’s the purpose of this book.

This book is about choosing... errr... umm... “good”. And achieving good with AI starts with mastering fundamental AI and data literacy skills. However, the foundation for those fundamental AI and data literacy skills is ethics. How we design, develop, deploy, and manage AI models must be founded on the basis of delivering meaningful, responsible, and ethical outcomes. So, what does ethics entail? I’ve tried to answer that in the next section.