Generative AI is the top IT spending priority for corporate executives, even though they lack trust that AI vendors will use their company data responsibly, according to a new survey from research firm IDC.
C-suite leaders’ second most pressing concern pertained to the accuracy of generative AI content, according to the survey of 895 C-suite leaders in North America, Western Europe, and the Asia Pacific regions.
“Trust and oversight are very important to the C-suite,” Teodora Siman, research manager of the C-Suite Tech Agenda at IDC, said during a webinar about the survey. “They want to understand how their data is being used, where it’s being housed.”
Tracking the Use of AI
Even before IDC conducted its survey, organizations began to voice concerns about employees sharing proprietary company information and other sensitive data with public generative AI tools, which may then use that data in responses to other users’ prompts.
To address this data security and privacy risk, ExtraHop in May released a new capability in Reveal(x) that gives organizations visibility into the devices and users on their networks connecting to generative AI domains, including ChatGPT, Google Bard, and GitHub Copilot.
By showing the users and devices connecting to generative AI services and the amount of data users are sending to generative AI domains, Reveal(x) enables organizations to determine whether sensitive data may be at risk of loss or exposure. The larger the volume of data employees are sending to generative AI domains, the higher the likelihood that sensitive data may be exposed. The capability gives organizations a mechanism to audit employee compliance with internal policies governing use of generative AI services.
Concerns About Accurate Results, Data Use
After launching its generative AI detection capabilities, ExtraHop also commissioned a survey to understand how company leaders plan to secure and govern the use of generative AI tools. The top concerns from the group of 1,200 IT and security decision makers were similar to those identified in the IDC survey.
In the ExtraHop study, respondents identified inaccurate or nonsensical results as their top concern about generative AI. Their second largest concern was exposure of personal information, followed by compliance violations for the exposure of personal data. Exposure of trade secrets or company intellectual property was among the top five concerns.
Meanwhile, in the IDC survey, the concern about AI vendors’ misuse of data extended to multiple types of company-held information, including intellectual property, confidential information, and customer and employee data.
Both surveys, however, found AI adoption high at large organizations. Just 13% of respondents in the IDC survey said they had no plans to adopt generative AI tools. Another 20% said they had already adopted generative AI, a third had begun implementing the technology, and another third were exploring how to use it.
Meanwhile, 73% of respondents to the ExtraHop study said employees in their companies use AI tools frequently or sometimes. In contrast, just 5% said employees in their organizations never use AI tools, even though 32% indicated that generative AI is banned at their workplaces. Presumably, if generative AI bans were effective, the number of respondents saying employees never use AI tools would be much higher than 5%.
Despite the gap between the percentage of companies that ban AI and actual employee use, 82% of those answering the ExtraHop survey said they were somewhat or very confident in their organizations’ ability to protect against AI threats.
More AI Investments Planned
Despite their relative confidence in their organizations’ ability to protect against AI threats, nearly three quarters of respondents in the ExtraHop study said they planned to invest in AI protections or security measures in 2023.
IDC asked which technology will be a spending priority over the next year, and 61% of security executives identified AI, compared with 69% of human resources executives and 57% of sales executives.
The IDC survey also asked executives about their overall IT spending in the next year. Just under half said they expect to increase IT spending, and nearly three in 10 said they expected their tech budgets to remain the same. Most of those expecting decreases in their IT budgets expected cuts of 10 percent or less.
Do you have concerns about generative AI? Join the conversation on the ExtraHop customer community.
1. Source: IDC Webinar, C-Suite Tech Agenda: Priorities & the Power of AI, October 18, 2023 https://goto.webcasts.com/starthere.jsp?ei=1634341