
So the theory goes that with MCP's, you can effectively monetize the value of your data that was trapped before. An alternative way of looking at it is that MCP's will instead commoditize data and make them harder to monetize, even data that is today perceived as proprietary. Mind that the data in itself is seldom valuable, just the distinctive business decision or unique workflow it may trigger. Consider the example of how financial markets have developed over the past century. In the past, trading data was opaque and proprietary and made it possible to take highly profitable trading positions with superior knowledge. As knowledge/data was standardized and distribution improved mostly thanks to tech innovation, the time to access the data then became the determinant of value - but even that eroded. Today, very few hedge funds have truly proprietary data and consequently the industry profits have decreased materially. So, in today´s AI-frenzied context, I would only invest in companies that have a near monopoly on their dataset over the long haul in order to make truly unique business decisions or design proprietary workflows to win in their market. Proprietary data isn't enough if all the industry peers are collecting their respective proprietary data sets as they will compete for the same customer decisions.









