Business leaders have high hopes that investments in artificial intelligence (AI) can lead to innovations that transform areas such as customer satisfaction and product development. However, inadequate data infrastructure is preventing 78% of organizations from achieving these goals.
A recent report from MIT Technology Review Insights, in collaboration with Snowflake, titled “Data Strategies for AI Leaders,” highlights that while many companies have high hopes for generative AI (72% are looking to increase efficiency or productivity, 55% want greater market competitiveness and 47% expect more innovation in products and services), it is crucial to improve data strategy to maximize the potential of AI.
AI in companies
Enterprises need a robust data infrastructure, powered by modern cloud platforms, that allows them to leverage both their own data storage and large volumes of previously inaccessible data, especially unstructured data such as videos and images. According to the study, only 22% of business leaders feel “very prepared” to adopt AI, while 53% consider themselves “somewhat prepared.”
Greater preparedness means facing fewer challenges related to access to scalable computing capacity, data silos and integration issues, and data governance. Despite many leaders’ confidence in the results AI can deliver, they are recognizing that data is key to determining how quickly and effectively they can unlock value from AI.
95% of organizations face challenges implementing AI
Another challenge for organizations is deploying AI on a large scale. 95% of respondents cited obstacles in implementing AI. 59% cited data governance, security or privacy as the most common challenge, followed by data quality and availability (53%) and cost of resources or investment (48%). Spending and resource allocation decisions, including those needed to improve data infrastructure, are challenging in any technology investment. However, the cost of generative AI is decreasing, as companies have begun to develop smaller large language models (LLMs) that offer the same capabilities at a lower cost.
“Many organizations today have high expectations for generative AI – they are looking to change how they operate and what they sell,” said Baris Gultekin, Head of AI at Snowflake. “Our joint research shows that as organizations feel greater urgency to deploy AI applications, they are realizing that their data can offer insights from previously untapped sources of information. “A robust data infrastructure is critical to delivering generative AI capabilities, and business leaders must act quickly to address issues such as data security and cost, and lay the foundation needed to deliver on the promise of AI.” .
The benefits of generative AI are increasingly visible to companies that are more advanced in their data strategy, as they have invested heavily in robust data infrastructures and are now being rewarded by applying AI to that data. Any company that wants to capitalize on AI must first establish a robust data infrastructure, encompassing a broad set of processes and assets involved in the collection, aggregation and storage of the organization’s data, as well as the ability to access it. Investing in robust data infrastructures across the organization will enable much more powerful generative AI users while reducing governance and security concerns.
