When an extra lane is suggested, NIMBY and/or environmental concerns will immediately raise objections and kill the proposal. So instead, the idea is to use the existing road in a more "efficient" way by making it "smart". This is of course ok, as it's not like we're adding any road, just using the existing road in a better way. After some time, accidents happen and people clamor for a hard shoulder, which should be ok environmentally as its not adding extra road, its just a hard shoulder for safety and not meant to drive on (and who can object to safety). And in the two step process, somehow we ended up with a widened road whilst completely avoiding the discussion if we want a wider road or not.
I sometimes also use this tactic in software if the PO's are non cooperative. First build the feature but hide it in a dev menu/behind a feature flag. No harm adding a feature if it's not available right? And then some time after it was implemented without explicit approval, I propose to enable the feature at zero cost, it is already implemented anyway.
You can also view the idea of smart roads in a more benign way of course. As cheaper way to experiment if adding an extra lane could be worth it. In the Netherlands it is done a lot, usually only during rush hour. It is then often paired with a stricter speed limit, so when the 4th lane is open the speed limit is 100km/h instead of 130km/h. But as the speed limit is 100km/h during day time anyways this is less of a change than it used to be.