With machine learning firms are continuously improving their products via customer data (often in real time). How much competitive advantage do firms get from this type of learning and is the outcome of such dynamic learning efficient?
In my recent paper with Andrei Hagiu, we call this type of learning "data-enabled learning" and model competition between firms that enjoy data-enabled learning. We explore the implications of the model for
competitive dynamics and efficiency. The model allows us to analyze factors affecting an incumbent's competitive advantage such as the shape of firms'
learning functions and the extent of data accumulation. We also explore the
implications of public policies towards data sharing, user privacy and data
acquisitions. Finally, we show conditions under which a consumer
coordination problem arises endogenously from data-enabled learning, thus
making it possible for consumer beliefs to matter for the incumbent's
competitive advantage and the efficiency of the outcome.
You can read the full paper here.
Commenti