Discover the Top Trends in Enterprise Machine Learning
In the past 12 months, there have been myriad developments in the field of AI/ML. Not only have we seen shifts in tooling, security, and governance needs for organizations, but we’ve also witnessed massive changes in the field due to the economic impacts of COVID-19.
Every year, we surveys business leaders and practitioners across the field for an annual report about the state of machine learning in the enterprise. This year was no different. In November, we conducted a survey with 400+ business leaders involved in AI/ML initiatives at their organizations, and discovered 10 key trends driving the industry in this year.
Here’s a preview of the key themes in the 10 trends that we discovered:
Theme #1: Organizations are increasing AI/ML budgets, staff, and use cases
A key takeaway from this year’s report is that enterprise IT departments are significantly increasing machine learning budgets and headcount for this year. Not only has the upheaval of 2020 not impeded AI/ML efforts that were already underway, but it appears to have actually accelerated those projects as well as new initiatives.
This year’s survey revealed that 83% of organizations have increased their budgets for AI/ML and that the average number of data scientists employed has increased 76% year-on-year. Organizations are also expanding into more use cases, especially in areas that can offer top- and bottom-line benefits during times of economic uncertainty.
Theme #2: Challenges span the ML lifecycle, especially with governance
Our report revealed that organizations are experiencing challenges across the ML lifecycle, but the top challenge by far is AI/ML governance. 56% of all organizations rank governance, security, and auditability issues as a concern, and 67% of all organizations report needing to comply with multiple regulations for their AI/ML.
Theme #3: Despite increased budgets and hiring, organizations are spending more time and resources—not less—on model deployment
Even though organizations are dramatically increasing their focus on AI/ML, they’re actually spending more time and resources on model deployment than they did before. Our report found that at 38% of organizations, data scientists spend more than 50% of their time on model deployment, and the time required to deploy a trained model to production has actually increased year-on-year.
What’s happening? Organizations seem to be using their increased headcounts to manually scale AI/ML efforts, rather than improving operational efficiency.
Theme #4: Organizations report improved outcomes with third-party MLOps solutions
The good news is, AI/ML is far more accessible than ever before and we believe organizations can overcome these issues. In the early days of AI/ML, any organization that wanted to deploy models at scale was essentially required to build and maintain their own machine learning operations (MLOps) infrastructure from scratch. However, organizations of all sizes can now get started with AI/ML more quickly and scale it efficiently.
Third-party MLOps solutions are one reason why the barrier to AI/ML is lower than ever before. And our survey found that organizations see improved outcomes when they use a third-party solution. Specifically, when compared to organizations that build and maintain their own systems from scratch, organizations that either integrate commercial point solutions into their systems or use a third-party platform spend less money on infrastructure, deploy models faster, and dedicate less of their data scientists’ time to model deployment.
