A function is provided:
\[ f(x) = x^2 + ax + b \]
A new function is defined:
\[ g(x) = f(x+1) - f(x-1) \]
\[ f(x+1) = (x+1)^2 + a(x+1) + b = x^2 + 2x + 1 + ax + a + b \] \[ f(x-1) = (x-1)^2 + a(x-1) + b = x^2 - 2x + 1 + ax - a + b \]
\[ g(x) = f(x+1) - f(x-1) \] \[ = [x^2 + 2x + 1 + ax + a + b] - [x^2 - 2x + 1 + ax - a + b] \] \[ = 4x + 2a \]
\[ g(20) = 4(20) + 2a = 80 + 2a = 72 \] This implies \( 2a = -8 \), so \( a = -4 \).
\[ f(x) = x^2 - 4x + b \]
Completing the square, we get: \( f(x) = (x - 2)^2 + (b - 4) \).
The term \( (x - 2)^2 \) is always non-negative. The minimum value of \( f(x) \) occurs at \( x = 2 \), which is: \[ f(2) = (2 - 2)^2 + (b - 4) = b - 4 \]
For \( f(x) \ge 0 \) to hold true for all \( x \), the minimum value must be non-negative: \[ b - 4 \ge 0 \Rightarrow b \ge 4 \]
The minimum possible value for \( b \) is \( \boxed{4} \).