The academic term (O'Reilly & Tushman) for an organization that runs explore and exploit simultaneously and well, rather than optimizing for one mode at the expense of the other.
Most companies default to a single management system — the KPIs, business plans, and reporting lines built for exploiting a known business — and apply it to exploratory work too, which is why exploration starves. Becoming ambidextrous means building separate but connected governance, teams, and evidence requirements for the two modes: a growth board with phase-appropriate evidence bars for explore, ordinary P&L management for exploit, and an explicit landing pad where proven explore bets cross over into the exploit side.
Osterwalder frames the failure mode simply: organizations are single-minded by default, and ambidexterity has to be designed in, not hoped for.