Choose maximal such that for the displayed , and take the OR of AND blocks of size , ignoring unused coordinates. Its failure probability is , between one quarter and three quarters. Active coordinates have influence and unused ones have zero influence. Maximality gives ; for , , so every influence is below . The construction is therefore a quite fair Boolean function with small influences for every . The logarithmic bound cannot hold at , where every quite fair function has influence one.
A Boolean function takes each vector of bits to either zero or one. Use independent uniform input bits; the influence of a variable is the probability that flipping that bit changes the value. A quite fair Boolean function has . All logarithms in the estimates below are natural, except where a base is explicitly written.
Here is a balanced tribes construction with unused coordinates valid for every . For each put , and choose the largest with . Partition of the bits into blocks of size . Let be one when at least one block is all ones, and ignore the other bits. This is a tribes function. With , its zero probability is .
For the function is a single bit and has mean . For , and . The elementary inequalities give
Thus both output probabilities are between one quarter and three quarters.
A used bit is pivotal exactly when the other bits of its block are one and every other block fails. Hence its influence is ; unused bits have zero influence. Maximality of gives . Also for : this is immediate for , while for one has . Therefore
This proves the requested existence with room in the constant. The dimension-one boundary case cannot satisfy the printed logarithmic bound: a quite fair function of one bit is nonconstant and has influence one, whereas . Thus the statement requires the usual nontrivial range (or its intended large- interpretation).