Nearly a year after its introduction, use of Microsoft 365 Copilot, Microsoft’s flagship generative AI tool for the office market, is growing rapidly within enterprises, with annual revenue expected to increase by billions of dollars However, many companies are still unsure about implementing it. To complete development.
That’s according to the latest semi-annual survey of technology executives by the CNBC Technology Executive Council, which polls CTOs and CISOs from companies ranging from retail to industrial, healthcare, energy and finance. A sampling of opinions from technology leaders including: service company.
Broadly speaking, virtually no companies are hesitant when it comes to implementing AI. Nearly two-thirds (63%) of technology executives surveyed say new investments in AI are accelerating, and the remaining one-third are evaluating their investments in AI. However, he says that his approach is cautious.
One of the most revealing findings of the CNBC survey was how quickly Microsoft’s Copilot, introduced about a year ago, is gaining dominance in enterprises, with 79 survey respondents % say their organization uses Copilot.
But the numbers for full and permanent implementation aren’t such a clear win for the tech giants, at least for now. Fifty percent of survey respondents said their organization had made the decision to roll out Copilot to all employees, while another 17% said they had not made the decision to fully adopt Copilot, and an additional 3 One in three companies say they are still in the testing phase.
When asked if it’s worth the cost ($30 per user per month for Copilot), tech leaders say they’re not sure. An equal number of respondents said yes or no to this question, but 50% of respondents said it was too early to know. . According to the CNBC survey, these numbers tend to increase over time, with only one-third of companies surveyed saying they have no plans to use Copilot in the future; Two-thirds say they plan to use Copilot specifically in the following areas: future.
Samsung Galaxy Book4 Edge laptop with Microsoft Copilot+ PC.
Bloomberg | Bloomberg | Getty Images
According to a recent Deloitte study, two-thirds of organizations are increasing their investment in next-generation AI based on “strong initial value.” However, if we zoom in on that value, a different picture emerges. Many enterprise AI projects remain in what Deloitte calls the pilot or proof-of-concept stage, with more than two-thirds of organizations saying they have moved less than 30% of their generational AI experiments into full production.
According to Deloitte, there are multiple reasons why companies remain unsure about enterprise AI adoption. Over 40% of companies struggle to define and measure the impact of their artificial intelligence initiatives, and while less than half have developed key performance indicators to effectively measure AI returns, many Companies say standard success metrics don’t work.
Deloitte research shows that many companies remain unsure that they have the infrastructure in place to reap the benefits of AI deployments, and some do not have the sufficient infrastructure needed to bring custom models into production. This includes concerns that organizations lack relevant data. Risk, regulation (e.g. in Europe) and governance issues are also weighing on companies to scale artificial intelligence more quickly.
How is the race for the AI generation different from Microsoft?
Microsoft, which keeps adoption data at a near-conclusive level, says the number of people using Copilot every day at work has nearly doubled from last quarter, and nearly 60% of Fortune 500 companies now use Copilot. They have publicly announced that they are doing so. The number of customers with more than 10,000 seats more than doubled from the previous quarter, including Capital Group, Disney, Dow, Kindryl and Novartis. At $30 per user per month, that’s $3.6 million per year for 10,000 employees.
Jim Rowan, head of AI and principal at Deloitte Consulting, said companies across the economy have learned from their experience with cloud adoption that total cost of ownership can quickly become high, and that experience has led companies to “overcommit.” He said that he is being cautious about this. To the vendor.
According to CNBC research, Microsoft is far ahead, but it’s not the only company with enterprise wins. Nearly two-thirds of enterprises use Azure cloud services from OpenAI, the largest AI partner and AI investor. Meta’s Llama large-scale language model is the next most popular generation of enterprise AI, used by 38% of companies, followed by Anthropic (13%), and many top Silicon Valley VC firms. Mistral, the French AI startup it supports, follows. So does former Microsoft and Google chairman Eric Schmidt, who uses it in 8% of companies.
A recent study by Morgan Stanley on Copilot found that 94% of CIOs expect to adopt Microsoft’s generative AI products over the next year, up from 63% in Q4 2023. 68% of CIOs cited Microsoft 365 Copilot. Azure OpenAI Services is next cited by 41% of CIOs.
Rowan said Microsoft’s lead in the enterprise AI race is surprising because the further companies move away from Microsoft 365 Copilot, the harder it may be for them to claim investment, at least in the early stages. No, he said. This tool fits into existing technology platforms that are already deployed across the enterprise and embedded in the way we work. But for many other AIs, Rowan says, “there’s not a very easy button to press like there is with Microsoft 365.”
With an enterprise-wide solution like Copilot, it’s still important to consider the specific training and engagement of your employees as much as the technology investment itself, or you’ll end up justifying the cost. Rowan says it will be difficult to do so. “There needs to be ongoing messaging around AI from executives about how they’re using it and how their employees are using it,” he said.
46% of CNBC survey respondents said at least half of employees in their organization are using AI. Additionally, 42% of respondents said a quarter of their workforce uses AI.
According to the survey, most organizations (84%) believe that some Gen AI tool suite already fits or will fit into their business plans.
At this point in the AI adoption curve, it is logically inconsistent for executives to say that their company needs to “move on to the AI journey” even though they don’t know what the KPIs or ROI will look like. It’s not a thing.
Rowan said the core approach to research and development should dictate thinking. There’s always a question of where companies should start, but not whether they should start their research and development efforts, he says. “Start with a set of use cases to get into AI value creation mode,” says Rowan. “Fail fast, make changes, and react.”
In his view, the alternative is worse. “Companies risk missing out on a moment when technology has advanced so fast that they are now behind the curve,” he said.
The semi-annual survey was conducted from October 8th to October 29th among a sample of CNBC Technology Executive Council members. twenty one.
