This from Jud Ready, the lead designer of the cell, in an email to Inside Greentech:

We are currently less than those values, but we have “only” been doing this for a little under a year. Right now we are @ ~7% when the sunlight strikes at 45° to the perpendicular (maximizes light trapping) – this is about double the “high noon” value of 3.5%. Planar cells decrease in performance as the sun-light angle changes away from high-noon – our increase due to the light trapping morphology and that is extremely novel I’m told. I hope to have > 40% before it is all over (due to light trapping, I can use thinner photoactive layers that means less chance of recombination.) FYI, other CNT-based PVs are 1-2% max.