Patents can provide insights regarding the state of the art of artificial Intelligence (AI) technology innovation, and thus, a strategic move of a company for the AI innovation leadership. To compare the technology innovation strategy of the three leading companies in the AI business, Google, IBM and Microsoft, patent information is exploited for the cross-competitor analysis.
According to Dr. Benjamin Gilad with the Academy of Competitive Intelligence, the goal of the cross-competitor analysis is to enable one to simplify predictions of competitors’ moves and countermoves when multiple competitors are involved. Through the cross-competitor analysis, one can understand that: Competitors’ behavior when there are several significant competitors; Paradigm shifts in industries undergoing rapid change or transition; Entry of new competitors into the competitive landscape; Future directions in strategic move among industry contenders. The strategic map is a tool used in the cross-competitor analysis for visualizing the competitive landscape: A chart for a strategic parameter No.1 (e.g. product/service portfolio, distribution channel etc.) vs. a strategic parameter No.2 (e.g. product/service price, quality, brand etc.).
To do the cross-competitor analysis for the AI, US issued patents of Google, IBM and Microsoft that are related to the AI are reviewed for identifying patents that cover the major AI technology innovations: machine learning, neural network, expert system, fuzzy logic and genetic algorithm, and AI Applications. Total of 1087 Google, IBM and Microsoft patents are selected for the AI cross-competitor analysis. The key AI patents of Google, IBM and Microsoft account about 10% of total key AI issued patents in the US as of 1Q 2016.
Following figure shows the Activity Index vs. technology innovations strategic map for Google, IBM and Microsoft. The size of the circle represents the total number of the patents for each technology innovation. Activity Index is a measure of a company’s relative technology innovation activities in a specific technology innovation field: Activity Index = share of a specific innovation sub-class in a company/share of a company’s patent in total patents, where share of a specific innovation sub-class in a company = patents (innovation sub-class)/patents (a company) and share of a company’s patent in total patents = patents (a company)/patents (total companies).
The map clearly shows each company’s competitive advantage in the major AI technology innovations. Google’s AI technology innovation is focused on the neural network. In essence, a neural network is an attempt to simulate the human brain. IBM’s AI technology innovation is focused on the genetic algorithm. A genetic algorithm is a search heuristic that mimics the process of natural evolution, which used to generate useful solutions to complex problems. Microsoft holds comparatively large patents in the machine learning. A machine learning refers to intelligent systems that can learn from data, rather than merely follow explicit programmed instructions.
Real men have fabs!