In the past, I’ve written extensibly about the enterprise artificial intelligence(AI) space analyzing some of the key vendors and platforms. Today, I would like to explore the different categories of technologies that are relevant in the enterprise AI market.
The raising popularity of AI technologies have created an incredibly crowded ecosystem of products and platforms. However, the number of categories on which enterprise AI products can be grouped remains relatively small. This article provides a taxonomy that covers the main types of AI products that are actively evaluated by enterprises. Similarly, if you are a startup or a venture capitalist in the enterprise AI space, most likely your products or technologies will fall into one of these categories. Let’s take a look.
5 Key Categories in the Enterprise AI Market
1 — AI Cloud Services
AI cognitive service cloud platforms are a popular choice for enterprise as they provide a relatively low entry point to leverage very sophisticated AI capabilities in areas such as vision, speech or natural language processing. Platforms such as Watson Developer Cloud, Microsoft Cognitive Services, Google AI Services or AWS AI APIs are dominating this category but there is plenty of room for startups that provide differentiated capabilities in the space.
2 — Enteprise ML Platforms
Platforms that enable the implementation and management of machine learning(ML) models are also gaining traction in the enterprise. These offerings go beyond the simple creation of ML models by providing robust infrastructure and toolset required to manage the lifecycle of ML solutions. Platforms such as H2O.ai as well as cloud services such as AWS ML, Azure ML or Google Cloud ML are relevant products in this area.
3 — Enterprise R Distributions
R remains one of the most popular languages for implementing advanced statistical and ML solutions in the enterprise. In that context, many enterprises have been able to nurture and acquire R engineering talent. As a result, R distributions are a relatively simple entry point into enterprise AI strategies. Microsoft R Server, H2O.ai and the popular R Studio are solid offerings in the space.
4 — Self-Service, Interactive Data Science Tools
Empowering data scientists with tools for authoring and creating AI models is a top priority in enterprise environments. While legacy tools such as SAS are still dominant in this space, there is an emerging ecosystem of platforms such as Jupyter, Zepelling or Google Cloud DataLab that are gaining traction in the enterprise.
5 — Deep Learning Frameworks
This might sound like a subcategory of the previous ones but the market opportunity is big enough that s deserves its own group. Enterprise distributions of deep learning frameworks such as TensorFlow, Caffe, Theano and others will be essential to streamline the adoption of deep learning capabilities in the enterprise. While cognitive cloud APIs are a great option for enabling well-known and relatively simple deep learning models, open frameworks are often required to implement custom models for more sophisticated scenarios. As a result, there is an interesting opportunity for startups that enable enterprise distributions of some of the most popular deep learning frameworks in the market.