We hear tons of conversation around the use of Artificial Intelligence (AI). We also hear that the application of AI might be the solution to almost everything. While that’s both an overstatement and unrealistic, there are a lot of companies and folks working to make AI do just that — solve everything.

But consider that AI and machine learning learn “how to think” and formulate ideas by accessing real data and information. How do we limit or eliminate biases that lie within those data? Do we edit or limit that data and bias? If we edit data, do we just inject a different bias?

We have seen examples where AI literally writes a different history than the one we know exists.

Those of us who lived it can tell if a retelling is different than the reality we lived. But folks younger have no frame of reference and might just accept what they are told.

So, as you use AI, incorporate it into more of your life and see it grow, consider the following questions:

1) Should there be controls on what AI is fed?
2) Without controls, will AI lead to a good or appropriate result (it will absorb biases, good and bad)?
3) If there are controls, who will control what the AI applications are fed? (interventions lead to different biases but not necessarily the correct ones)
4) Who edits the biases absorbed into the AI applications? (and why are they the ones to do it?)
5) Who checks or oversees or analyzes the AI output?
6) Do we accept AI results as fact?
7) What happens if we accept the AI results, but they are wrong? Or proven wrong after use or action?
8) Do we limit the uses of AI until we know what it is and how it learns? (and does this limit its evolution?)
9) The government seems intent on regulating AI. How can we be confident this will fix the challenges? Should individuals be more actively involved in these ideas and conversations?

I naturally question things and look at how they can be made better…more effective and efficient. I worry a reliance – or even just an acceptance – of AI may crush that same tendency in others, particularly younger folks. We need to foster public will power to persevere a spirit of questioning. That’s the true path to solutions, not blind faith in AI.

This blog is part of a series on AI and its impact on policy, health care and life. Read Errors in formulas and what it means for AI and, coming later this month, Does AI need guardrails?