Many CIOs are stuck in the AI slow lane but setting ambitious priorities as if they’re not

By John KellContributing Writer and author of CIO Intelligence
John KellContributing Writer and author of CIO Intelligence

    John Kell is a contributing writer for Fortune and author of Fortune’s CIO Intelligence newsletter.

    CIOs are setting AI priorities as if they're in a race.
    CIOs are setting AI priorities as if they're in a race.
    Rich Graessle—Icon/Getty Images

    Many chief information officers are living in conflicting realities when it comes to generative AI. Their top priority for the buzzy technology this year—creating new revenue streams—comes despite the modest progress they’ve made in doing anything useful with it.

    Only 26% of CIOs have deployed AI in any meaningful way, according to a new survey from consulting giant EY. And less than half have even got to the point of a test project using AI or proof of concept.

    The survey of 500 U.S.-based CIOs at firms with more than $1 billion in annual revenue, in partnership with advisory firm Oxford Economics, found that 37% of CIOs say creating new revenue streams was their most important objective for generative AI deployment, followed by reinventing their businesses at 27%. Only one in five say their main reason for deploying AI is reducing costs.

    And yet most generative AI adoption up to this point has been focused on back-office tasks, like customer service chatbots, AI copilots, or speeding up coding. It’s much easier to automate work that employees do today than reinvent what may be possible tomorrow.

    “If we’re coming off a year where 2023 was all about experimentation and a year of learning, I think the ambition for many CIOs is starting to shift towards revenue generation and growth,” says Dan Diasio, global AI consulting leader at EY. “But the findings still suggest that people are focused on automation, as opposed to transformation.”

    In short, there’s a big disconnect between the projects CIOs are doing and those they have set as their priorities.

    This reminds me of a recent conversation I had with Chris Bedi, chief digital information officer at business software provider ServiceNow. He’s trying to transform the company’s workforce in two phases. “Mode one” is about incremental improvements to make work a bit easier, offloading some repetitive or time consuming tasks to AI. “Mode two,” which will take longer, is answering the question: How would every job at ServiceNow be completely rebuilt from scratch with an AI-first mindset, in which certain tasks are completely handed over to technology?

    “If you zoom out and fast forward 12 to 18 months, it is going to become unthinkable for people not to have gen AI infused in their work,” says Bedi.

    Diasio agreed that AI will revamp how people do their jobs. “The true value of generative AI is not in doing our work a little bit faster, but changing the way we do our work,” he says. 

    Last year, CIOs and other C-suite leaders were figuring out the infrastructure needed to implement generative AI by compiling a list of dozens of potential use cases, determining budget priorities, and reorganizing their teams.

    Diasio suspects many CIOs have been too scattered with their AI playbooks. Large companies are toying with dozens of projects that would infuse generative AI into their business, aiming to transform nearly every job function. But for many, that everything approach has led to paralysis in terms of actually deploying AI.

    Instead, it may make sense to focus on fewer projects, particularly aimed at internal use rather than for customers, such as an AI tool that can help employees with HR questions or IT problems.

    “Being able to stand something up in production for use internally within organizations, that’s a major milestone for CIOs to say they’ve been through the lifecycle at least once and they’ve learned from that,” says Diasio. After deploying AI, CIOs can get practical experience with data management, cybersecurity, and the costs of AI that can help them navigate other projects.

    When generative AI programs are led jointly between the CIO and chief executives, EY says there’s more potential for companies to place bets on AI outside of IT, where the technology may actually disrupt the status quo. Most CIOs are dedicating at least 10% of their annual IT budget to generative AI, EY says. But when a CIO and CEO jointly lead the AI agenda, generative AI investments rise to an average of 13% of the total IT budget. 

    “If it’s just the CIO that owns the agenda, it’s difficult to get the level of authority to be able to make that change happen,” says Diasio. “Where we see the power of the CIO working with the CEO is in helping to build and place the bets outside of the IT organization, where generative AI is going to disrupt or reinvent the business.”

    John Kell

    Send thoughts or suggestions to CIO Intelligence here.

    NEWS PACKETS

    OpenAI debuts a new faster, cheaper AI model. A new updated version of OpenAI’s GPT-4 model, GPT-4o, debuted on Monday and will be better at handling text, audio, and images in real-time. The updates, which will be available in the coming weeks, will also make certain features free to all users that had previously been for paid subscribers to ChatGPT, Bloomberg reports. The update was opportunistically planned a day before the Google I/O developer conference, where, among other things, it unveiled a new feature called AI Overviews that will show AI-powered answers to billions of people using its namesake search engine by the end of this year. 

    Anthropic launches generative AI chatbot in Europe. The Alphabet-backed startup has released Claude, a direct competitor to ChatGPT, via the web and iPhones across the European Union and a handful of non-EU nations. CEO Dario Amodei says Claude was designed with “a strong commitment to accuracy, security and privacy,” perhaps a nod to the more stringent regulatory climate in Europe. Earlier this year, the EU Parliament passed the world’s first major set of regulatory rules to govern AI.

    But Anthropic’s Claude is trailing ChatGPT. Earlier this month, Anthropic launched its first mobile app on iOS and pulled in 157,000 total global downloads in its first week, badly trailing the 480,000 downloads for ChatGPT in the first five days of its U.S. launch when the app was also only available on iOS, TechCrunch reports. Demand is strongest in three markets: the U.S., Japan, and the U.K. But data from app intelligence firm Appfigures also shows that interest in Claude fell off steeply throughout the first week of launch, an indication that brand recognition isn’t as strong as ChatGPT.

    Apple plans to bring generative AI to iPhones. At Apple’s annual developers conference in June, the tech giant is expected to release an improved Siri that will be more conversational and feature a new generative AI system that will let it chat rather than respond to questions one at a time. Siri’s update reflects an acknowledgement by Apple executives that the virtual assistant looked antiquated when tested against ChatGPT, the New York Times reports.

    ADOPTION CURVE

    Four out of 10 CIOs say they don't expect to see a return on investment on their AI spending for at least the next two years, though on the flip side, another 39% saying they are already seeing positive ROI, according to a survey of 750 IT leaders conducted by PC maker Lenovo.

    The survey also found while 96% of CIOs anticipate spending more on AI over the next 12 months, 61% admit they find it “very challenging to demonstrate ROI with tech investments.”

    JOBS RADAR

    Hiring:

    - Centers for Medicare & Medicaid Services, a federal agency that provides health coverage to more than 100 million people, is seeking a CIO. Posted salary range: $147.6K to $221.9K/year.

    - The City of Cleveland is seeking a chief innovation and technology officer. Posted salary range: $160K to $185K/year.

    Hired:

    - Inspire Brands announced that Yasir Anwar has joined the restaurant operator as CTO, after previously serving as chief technology and digital officer at Williams-Sonoma. In his new role, he will oversee several teams, including product engineering, cybersecurity, and technology operations.

    - Freshworks named Ashwin Ballal as CIO, overseeing IT infrastructure, systems, and operations. Most recently, Ballal was EVP and CIO at Medallia.

    - Cohesity announced the appointment of Dr. Craig Martell as chief technology officer. He most recently served as the chief digital and AI officer for the Department of Defense.

    - DBS Group Holdings has named Eugene Huang as CIO, effective May 10. With over 36 years of tech and banking expertise, Huang joins the Singapore-based financial services firm from Ping An Group, where he led a team of 22,000 employees providing IT offerings including infrastructure and cybersecurity. 

    - iCIMS has appointed Joseph Benjamin as CTO, effective this week, where he will help lead the technology and product teams alongside Eric Connors, who joined the company earlier this year as chief product officer. Both will also be responsible for accelerating iCIMS’ AI program.

    - R&T Deposit Solutions announced the appointment of Mieko Shibata as CIO, where she will lead the firm’s technology strategy and serve as a member of the executive management committee. Shibata comes to R&T from Bank of America, where she held leadership roles and served as managing director.

    - The Cardano Foundation, a Swiss-based not-for-profit, appointed Giorgio Zinetti as CTO. He is the first to take on the role and will help increase Cardano’s adoption of blockchain solutions.

    This is the web version of CIO Intelligence, a weekly newsletter on the tech, trends, and news IT leaders need to know. Sign up for free.