Introduction

451 Research’s Voice of the Enterprise: AI & Machine Learning survey provides a vital quantitative window into the current state of enterprise AI. The newest iteration of this survey, the recently published Use Cases 2020 report, contains metrics for important aspects of AI and machine learning, including adoption rate and maturity, enterprise strategy, use cases, vendors, benefits and barriers, and decision-makers. This wave was completed in November 2019 and includes data from over 1,000 respondents across North America and Europe.

The 451 Take

Even at this early stage of enterprise AI adoption, the market is showing a clear trend. Almost half of AI adopters identify cloud-based AI platforms as their primary AI adoption strategy, and almost all plan to use one within three years. This prevalence is the result of two elements these products bring to the table: core technological capabilities necessary to build cutting-edge AI applications and the ability to customize component pieces for customers’ unique needs. While there are several vendors in this space, many different qualities are driving purchasing decisions of customers, making this market one to watch.
 

Primacy of cloud-based AI platforms

As we have noted in previous research, cloud-based AI platforms are an important component of enterprises’ AI adoption strategies, and new data suggests these platforms have become even more integral. As demonstrated in Figure 1 below, 48% of AI adopters identify these services as their primary AI adoption strategy, ahead of AI-enhanced applications (25%), open source tools (17%) and systems integrators (11%). These numbers are dramatically different from previous results, in which 38% cited AI-enhanced applications as their primary strategy and only 24% said the same of cloud-based AI applications.



Why are AI adopters centering on cloud-based AI platforms as their go-to strategy? Simply put, the offerings from these vendors provide the best of both worlds: just the right combination of fundamental tooling and customization needed to bring successful AI initiatives to fruition. There are many steps to building an AI application, from dataset preparation to training and inferencing – and all that is before you get to more traditional areas such as workflows and user experience. Plus, it is a continuously iterative process. By abstracting or automating some or parts of these steps, cloud-based AI platforms let data scientists and machine learning experts use their time more impactfully.

Cloud-based AI platforms also align with the increasingly common cloud-native approach to IT infrastructure. The technologies underpinning these products – containers, microservices, service meshes, etc. – beget the flexibility and resilience necessary for large-scale deployment of automated decision-making systems driven by machine learning application.

The Vendor Landscape

The vendor landscape in regard to cloud-based AI platforms, which has remained essentially unchanged in year-over-year data, can be divided into three tiers:

  • Public cloud providers. Microsoft Azure, Amazon Web Services, IBM and Google Cloud Platform are leading the AI platform market. These providers have recognized the importance of AI capabilities both in terms of driving customers to their platforms and encouraging them to increase their usage. In addition to building out capabilities within core AI functions like computer vision and natural language processing, these companies have recently expanded features around machine learning operationalization and even moved into horizontal applications such as contact center software.
  • Software vendors. Companies like Oracle, Adobe, SAP and Salesforce have been offering enterprises specialized software products for decades, and they already have vast volumes of enterprise data in these systems. From their perspective, it is important not only to infuse machine learning into their applications, but also allow customers to train and deploy their own applications using their own data. To this end, like the public cloud providers, these vendors have built out machine learning capabilities on top of their platforms.
  • Pure-play vendors. There are several providers in this market – such as DataRobot, H2O.ai and C3.ai – whose primary products are cloud-based AI platforms. While these companies are relatively new to the fold, they often differentiate themselves through advanced functionality and open architecture. In some cases, they offer vertical-specific applications to help customers get up and running in niche use cases.

Factors Driving Vendor Selection

It is important to point out that this is not a winner-takes-all market because the average adopter of cloud-based AI platforms leverages multiple vendors. At this early stage of adoption, there are enough non-overlapping capabilities and distinct functionalities to justify splurging on the services of more than one vendor.

For the first time in this survey, we collected data on what qualities and features are driving selection of cloud-based AI platforms vendors. Top of the list is security features, which 38% of adopters cited as a key driver of purchasing decision-making for these products. The prominence of security concerns has been a key theme in other areas of AI – it was the most commonly cited concern for enterprises about their AI infrastructure. Enterprise demand for security can be somewhat reflexive, but it is an important part of AI systems, which handle extensive amounts of sensitive data.

Another interesting takeaway is the secondary status of machine-learning-specific features. The top seven qualities – covering elements like cost, ease of use and data management – are components of any data management and analytics platform, if not software offerings in general. The top criterion specific to AI and machine learning is model deployment tools, cited by 23% of adopters. In our eyes, this suggests we are still at a nascent stage of AI adoption where advanced functionality is not driving vendor selection. 
Nick Patience
Founder & Research Vice President

Nick Patience is 451 Research’s lead analyst for AI and machine learning, an area he has been researching since 2001. He is part of the company’s Data, AI & Analytics research channel but also works across the entire research team to uncover and understand use cases for machine learning.

Jeremy Korn
Associate Analyst

Jeremy Korn is an Associate Analyst for the Data, AI & Analytics Channel at 451 Research, where he covers artificial intelligence and machine learning in the enterprise. In particular, he focuses on the legal and ethical challenges raised by these emerging technologies. In addition, Jeremy helps lead the Voice of the Enterprise: AI and Machine Learning survey, which provides qualitative insights into AI adoption, use cases and infrastructure.

Keith Dawson
Principal Analyst

Keith Dawson is a principal analyst in 451 Research's Customer Experience & Commerce practice, primarily covering marketing technology. Keith has been covering the intersection of communications and enterprise software for 25 years, mainly looking at how to influence and optimize the customer experience.

Want to read more? Request a trial now.