A survey of AI in large companies sees boom times ahead — if AI teams can get past issues that surround implementation.
What’s new: Businesses of all sizes are using more machine learning, spending more on it, and hiring more engineers to wrangle it, according to a survey of 750 business leaders by Algorithmia, which provides tools that automate model deployment and management. Nonetheless, struggles with deployment, scaling, and other issues continue to hinder adoption.
What they found: The survey questioned executives in a variety of sectors including finance, healthcare, education, and information technology. More than two-thirds of those who responded said their AI budgets are growing, while only 2 percent are cutting back.
- 40 percent of companies surveyed employed more than 10 data scientists, double the rate in 2018, when Algorithmia conducted its previous study. 3 percent employed more than 1,000 data scientists.
- Many respondents said they’re in the early stages, such as evaluating use cases and developing models.
- Many struggle with deployment. Half of those surveyed took between 8 days and three months to deploy a model. 5 percent took a year or more. Generally, larger companies took longer to deploy models, but the authors suggest that more mature machine learning teams were able to move faster.
- Scaling models is the biggest impediment, cited by 43 percent of respondents. In larger organizations, this may reflect siloing of machine learning teams in various departments. The authors believe that the solution is to centralize AI efforts in an innovation hub like those launched by Ericsson, IBM, and Pfizer.
Behind the news: Several other recent surveys shed light on AI’s evolving role in the business world. For instance, MIT Technology Review looked at AI’s growth in different global regions, and McKinsey examined how different market sectors, like manufacturing, marketing, and supply chain management, are finding profitable uses for the technology.
Why it matters: AI is new enough, and evolving fast enough, that every company’s experience is different. Spotting areas where industries where machine learning is having an impact, as well as trouble spots in deployment, can help guide crucial decisions.
We’re thinking: In 2019, many companies experimented with AI. In 2020, a growing number started talking about how to productionize models. In the coming year, we hope for rapid progress in MLOps processes and tools to make building and productionizing machine learning systems repeatable and systematic. AI Fund (where Andrew is managing general partner) has seen a lot of startups jump into this space, which bodes well for the future.