We know what we already know. We also know that our knowledge includes both accurate and inaccurate information. An example of accurate knowledge: the polio vaccine, introduced in 1950, has since reduced the incidence of paralytic polio globally. An example of inaccurate knowledge: standard childhood vaccinations cause autism, or mental retardation.
We don't know what we don't know. What we don't know includes known unknowns and unknown knowns. Examples of known unknowns: We know that one country will prevail in the World Cup, but it hasn't been decided yet. We know that the Asteroid belt between Mars and Jupiter contains big rocks that pose a threat to our survival, we just don't know precisely which rocks, or when or where they'll impact Earth. A couple explamples of unknown knowns: The true reasons for invading Iraq in 2003 were intentionally kept unknown, except to a select few, hidden behind the false reasons of WMD. A better means for identifying asteroids that threaten our survival may already exist, in the form of an unfinished algorithm on some astro-physicist's laptop.
We don't know what we will know. By definition, we can't know what we will know, or we'd know it already. And yet security experts, financial consultants, media pundits, and even Secretaries of Defense, behave as if they do know what they will know. Predictably, they predict events based solely on knowledge of what they already know. It's an invariably "safe" bet that doesn't cause consternation or disapproval because it's what everyone expects. Yet in retrospect, this method of predicting consistently produces unintended, even disastrous, consequences (see: Iraq, Vietnam, the Internet bubble, the housing bubble, etc.). What we will know cannot truly be ascertained, but it may be glimpsed -- if we continuously challenge what we know, and re-examine what we don't know.