survey

A survey of 800 IT and business leaders finds only 37% believe their infrastructure and data ecosystem is well-prepared for implementing generative artificial intelligence (GenAI).

Conducted by Enterprise Strategy Group (ESG) on behalf of Hitachi Vantara, the survey also finds that 40% of respondents are not well-informed regarding planning and execution of GenAI projects.

Nearly two-thirds (63%) of respondents, however, report their organizations have already identified at least one use case for GenAI, with the most cited use cases centered around process automation and optimization (37%), predictive analytics (36%) and fraud detection (35%). Nearly all respondents (97%) said GenAI is among their organization’s top five priorities but only 43% report having realized benefits up to this point.

A full 96% also said their organization prefers an alternative to proprietary models, However, in terms of the tools and technologies used to work with large language models (LLMs) only 23% said they prefer to mainly work with open source tools. That compares to 24% prefer to predominately use proprietary tools they developed. Just under a third are relying mainly on commercial tools, while 20% are using a mix of commercial and open source tools.

Additionally, 86% said their organization is making use of retrieval-augmented generation (RAG) to some extent to expose enterprise data to an LLM.

Finally, more than three quarters (78%) said their organization is using a mix of on premises and public cloud for building and using GenAI solutions, but 71% also noted their infrastructure needs to be modernized before pursuing GenAI projects, the survey finds. More than three quarters (77%) said data quality issues needed to be addressed before accepting the results of GenAI outputs.
Overall, the top generative AI adoption challenges are security (38%), cost and technical debt (27%) and data availability and quality (27%), the survey finds.

Ultimately, most of the IT infrastructure used for the inference engines required to run GenAI applications will be managed by internal IT operation teams alongside existing server and storage platforms or consumed via some type of as-a-service platform, says Jason Hardy, Hitachi Vantara CTO for AI.

In many cases, where a generative AI application is deployed is going to be predetermine by where data is already located simply because the cost of moving data is too high, he adds. “Data has gravity,” notes Hardy.

However, in the short term more than half of survey respondents (53%) said their IT teams are unable to keep up with the pace of innovation being driven by GenAI. A total of 61% added there is significant room to improve collaboration among teams responsible for selecting, implementing and managing GenAI solutions and infrastructure.

Less than half (44%) of organizations have well-defined and comprehensive policies regarding GenAI, the survey finds. A full 81% have data privacy and compliance concerns as well.

One way or another, the rise of AI will inevitably drive a refresh of IT infrastructure that in many organizations is long overdue. The only thing left to determine is how funding for that infrastructure will be attained, where it is going to be deployed and who is going to be responsible for managing it.

TECHSTRONG TV

Click full-screen to enable volume control
Watch latest episodes and shows

AI Field Day

TECHSTRONG AI PODCAST

SHARE THIS STORY