Algorithmia Report Reveals Top Enterprise Trends in Artificial Intelligence/Machine Learning for 2021
The last couple of years have witnessed major growth in the field of Artificial Intelligence (AI) and Machine Learning (ML). The proliferation of the internet, smart handheld devices, advanced computing hardware, and loads of user-generated data have been the driving factors behind this growth. We have seen major shifts in tools, security, and governance needs for organizations, especially, due to the drastic changes and the economic impact caused by the coronavirus pandemic. As such, companies have turned their attention towards AI-powered solutions capable of delivering both short-term and long-term technology innovation to drive revenue and efficiency in these uncertain times. Consequently, the efforts related to AI/ML have increased with enterprises significantly increasing their budget and teams for 2021. Perceiving the situation, Algorithmia, a leader in ML operations and management software, has published its 2021 Enterprise Trends in Machine Learning report, outlining the priorities and challenges of enterprise IT departments pursuing AI/ML initiatives.
The blind study, which included 403 business leaders involved in machine learning initiatives at companies with $100M or more in revenue, reveals that enterprise IT departments are increasing machine learning budgets and headcounts, despite not learning how to translate increasing investments into efficiency and scale.
Recommended AI News: Givenchy Tested « Vertical Video », Dailymotion’s New Mobile Ad Format
Here are some of the trends that come out of the report.
#1. Organizations Are Increasing AI/ML Budgets, Staff, and Use Cases
Algorithmia’s 2020 report highlighted that organizations were already increasing their investments in AI/ML before the pandemic. However, the emergence of the COVID-19 sped up the process. The 2021 survey revealed that 83% of organizations have increased their budgets for AI/ML and that the average number of data scientists employed has increased by 76% year-on-year. In addition, it highlights that enterprises are adopting AI/ML to a wider range of use cases. The percentage of organizations that have more than five use cases for AI/ML has increased by 74% year-on-year. The top areas of focus are related to the customer experience and workflow automation which are recognized as the top contenders for bringing in benefits during times of economic uncertainty.
#2. Challenges Span the ML Lifecycle, Especially with Governance
The major challenge faced by organizations in implementing AI/ML models is its governance. Around 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. Additionally, corporations continue to struggle with the basic deployment and organizational challenges. While the cross-functional alignment continues to be a major challenge to achieve AI/ML maturity, 49% of organizations ranked basic integration issues as a concern.
#3. Despite Increased Budgets and Hiring, Organizations Are Spending More Time and Resources—Not Less—on Model Deployment
Despite having additional budgets and headcounts, organizations are spending more time and resources on model deployments than they did before. According to Algorithmia’s report, the time required to deploy a trained model to production increased year-on-year, and 64% of all organizations take a month or longer to deploy a model. While 38% of all organizations are spending more than 50% of their data scientists’ time on model deployment—and organizations with more models spend more of their data scientists’ time on deployment, not less. All-in-all, the companies have increased resources without solving underlying challenges with operational efficiency, which has led companies to spend more time and resources on model deployment.
Recommended AI News: Ahana Announces General Availability of Managed Service for Presto on AWS
#4. Organizations Report Improved Outcomes with Third-party MLOps Solutions
Algorithmia’s 2021 report highlights that organizations see improved outcomes when they use a third-party solution to manage their machine learning operations (MLOps). Companies that either integrate commercial point solutions into their systems or use a third-party platform spend an average of 19-21% less on infrastructure costs when contrasted with companies that build and maintain their own systems from scratch. On average, their data scientists also spend a smaller portion of their time on model deployment and it takes them less time to put a trained model into production.
“COVID-19 has caused rapid change which has challenged our assumptions in many areas. In this rapidly changing environment, organizations are rethinking their investments and seeing the importance of AI/ML to drive revenue and efficiency during uncertain times,” said Diego Oppenheimer, CEO of Algorithmia. “Before the pandemic, the top concern for organizations pursuing AI/ML initiatives was a lack of skilled in-house talent. Today, organizations are worrying more about how to get ML models into production faster and how to ensure their performance over time. While we don’t want to marginalize these issues, I am encouraged by the fact that the type of challenges have more to do with how to maximize the value of AI/ML investments as opposed to whether or not a company can pursue them at all.”
You can read all the trends in Algorithmia’s 2021 Enterprise Trends in Machine Learning report here.
Recommended AI News: Grain Connect Teams with DZS to Expand Gigabit-Speed Fiber Network Throughout U.K.