-
Early adoption of AI has been riddled with concerns around data privacy, governance, hallucination and security
-
A group of 51 engineers, product builders, data scientists, enterprise leaders and startup founders said that when it comes to AI development, 53% consider safety a priority over innovation
-
Migration costs — from on-prem to cloud, and now AI adoption — are a “tough pill to swallow,” according to the head of product at AI-enabled team formation platform A.Team
Enterprises are on the move when it comes to AI adoption, with recent polling from Gartner finding that over half (55%) of organizations have increased investment in generative AI (GenAI) within the last 10 months — that is to say since it became the elected tech theme of the year following ChatGPT’s public release late last year. Still, that figure was a sharp increase from Gartner’s polling conducted in the spring of 2023, where only 19% indicated increased investments.
But during a webinar hosted by data privacy and analytics company TONIC and member-based consultancy firm A.Team, a group of 51 engineers, product builders, data scientists, enterprise leaders and startup founders took a live poll revealing that when it comes to AI development, 53% consider safety a priority over innovation while a contending 47% felt innovation should take precedence.
A.Team's head of product, Matan-Paul Shetrit, told Silverlinings the team was surprised at how close the polling was. “My guess is a lot of the people that say we need to prioritize innovation [are] probably people who have seen what happens if their companies don't act quickly,” he noted — perhaps learning from prior industry moves to cloud or mobile.
Shetrit likens the ever-rapidly-evolving space of tech to building a race car while driving at 200 miles an hour, and AI (particularly GenAI) is only ramping up the speeds.
Adoption concerns
Early adoption of AI has been riddled with concerns around data privacy, governance, hallucination and even the security of the AI engines themselves. In fact, A.Team’s polling found that 89% of respondents are concerned about their data being used to train large language models, prompting the release of vendor tools like ChatGPT’s enterprise-focused service.
But just as AI itself is not a new technology — with ML being “the less sexy GenAI experience” — many of the problems and needed guardrails have been around for some time. From Shetrit’s point of view, the solution for many of these issues looks a lot like a cloud vendor or cloud-based service, a model that “runs on your data, your clusters, it's fully encapsulated in partition, and there's no cross-pollination.” He is seeing a “clear understanding” in the industry when it comes to making that happen. “Are the concerns legitimate? Yes. Do I think they're solvable? Absolutely,” he said.
Shetrit also acknowledges that migration costs — from on-prem to cloud, and now AI adoption — are a “tough pill to swallow,” especially when trying to stay in the race at 200 miles per hour. His team’s focus has been on helping organizations take a more cost-effective, targeted approach to this migration through the use of fractional teams, he explained.
“Instead of trying to reinvent the wheel, they’re coming in with that domain expertise and helping you push forward these strategic initiatives," he said.
Want to discuss AI workloads, automation and data center physical infrastructure challenges with us? Meet us in Sonoma, Calif., from Dec. 6-7 for our Cloud Executive Summit. You won't be sorry.