Do-it-yourself AI projects are alive and well as CxOs are trying multiple approaches for generative AI, getting some projects to production and looking for better returns, according to a pop-up attendee survey at Constellation Research's AI Forum.
In a survey of 35 CXOs at the Constellation Research AI Forum, it's clear that the genAI playbook is far from being solidified. Forty-three percent of respondents were from companies with revenue of more than $1 billion. Overall, the pop-up survey at AI Forum aimed to highlight what CxOs were doing directionally. Salesforce CEO Marc Benioff said do-it-yourself genAI doesn't make sense in the long run because there's too much work involved and most enterprises won't keep up. That argument has a lot of merit, but it remains to be seen how genAI projects play out. T-Mobile and OpenAI announced they were building custom applications in a partnership.
Respondents indicated that they were using multiple approaches to build AI capabilities. The majority (79%) said they were developing home-grown AI services on hyperscale cloud services and 48% were also using open-source frameworks and large language models. Many of these efforts included AI embedded in packaged applications that they already used such as Salesforce, Adobe, Oracle, SAP etc.
Automation was the biggest reason to implement AI with operational efficiency No. 3. The No. 2 reason for implementing AI was for cutting edge capabilities.
More from AI Forum:
- Sam's Club CEO Nicholas on AI, frictionless commerce, focus on members
- Boomi CEO Lucas: AI agents will outnumber your human employees soon
ROI, however, was a bit elusive. Forty-five percent of respondents said they have yet to see ROI from AI technology and 31% said they've deployed with models returns. Of the respondents implementing AI, all of them said there was room for improvement whether they were seeing ROI or not. The majority of respondents all said their AI investment in 2025 would be up.
Investment priorities included data lakes, predictive analytics, natural language processing and image recognition.
As for functions, CxOs at AI Forum said they have scaled AI projects in employee productivity, back office, IT and sales and marketing.
Other items of note:
- 40% of CxOs at the AI Forum noted they didn't have the human capital to successfully implement AI.
- CxOs were recruiting, internal peer networking, training existing employees and partnering with companies and universities to fill talent gaps.
- Roles are changing as managers aim to acquire data expertise and restructure business models. Managers are also networking to gain knowledge.
- CEOs, CTOs and CIOs are generally leading the AI charge.
- Operational efficiency and revenue and growth are areas driving the most ROI for AI projects.
- Trust, budget and data quality are the three challenges limiting ROI.
- OpenAI, Llama and Anthropic models were most popular in order.
