According to a report this week from Business Insider, Meta has decided to give its employees access to a range of different AI tools, including those made by its competitors in the AI space: Google, Anthropic and OpenAI. Instead of restricting employee use to its own large language model (LLM) known as Llama, Meta has eliminated barriers in its mission to make its workforce “AI-first.”
In practice, this means employees now have authorized, paid access to a selection of the latest and greatest tools in generative AI, some of which are likely already personal favorites of many Meta staff.
But opening the floodgates to multiple AI providers and tools does not ensure effective adoption. For CIOs, deciding which AI tools to roll out is just the first step in securing ROI. When investing millions into new technology, making sure that the AI toolkit actually supports and engages employees is critical — and requires comprehensive education. Offering more options could help improve the chance that workers will find something useful for their workflows, but CIOs can’t rely on that alone.
“At this point, AI adoption isn’t a technology issue — it’s an operating model issue,” said Patrice Williams Lindo, workforce futurist and founder of Built Different Conference. “The companies pulling ahead are the ones aligning IT governance with people strategy, instead of forcing employees to navigate the gap alone.”
AI dreams vs. reality
After multiple years of relentless hype around AI and its promises, it’s no surprise that companies have high expectations for their AI investments. But the measurable results have left a lot to be desired, with studies repeatedly showing most organizations aren’t seeing the ROI they’d hoped for; in a Deloitte research report from October, only 10% of 1,854 respondents using agentic AI said they were realizing significant ROI on that investment, despite 85% increasing their spend on AI over the last 12 months.
This disconnect between financial investment into AI and its material gains stems from several different issues — which can then often exacerbate each other.
“We’re throwing AI out there and seeing what sticks on the wall,” said Beverly Weed-Schertzer, author and executive consultant for IT education and training at edifyIT and global program manager at BT. “But it’s still technology — and like anything else, there has to be training and education.”
Too often, a company chooses an AI tool that seems useful and exciting but doesn’t clearly translate to employee needs. Weed-Schertzer weighted the importance of picking the right AI tool at just 35%, with 65% coming down to effective process and people management. Without a useful example of implementation, employee adoption rates stagnate, and the effectiveness of the AI deployment is restricted — even if it technically works perfectly.
Williams-Lindo agreed that many companies are struggling to formulate effective AI strategy and emphasized that failed ROI can’t be attributed to employees themselves. Instead, it should be placed at the feet of leadership: “AI adoption isn’t failing because workers aren’t ready. It’s failing because leadership hasn’t decided what kind of organization it wants to be in an AI-enabled world,” she said.
Certainly, there is no point in spending millions on a toolkit if it doesn’t align to meaningful application within the enterprise. But whose responsibility is it to figure out effective implementation? Perhaps surprisingly, the experts all agreed: it’s not just the CIO.
Who owns AI implementation and adoption?
At face value, it seems obvious that the IT leadership team should be responsible for all things AI, since it is a technical product deployed at scale. In practice, this approach creates unnecessary hurdles to effective adoption, isolating technical decision-making from daily department workflows. And since many AI deployments are focused on equipping the workforce with new capabilities, excluding the human resources department is likely to constrain the effort.
“AI exposes a long-standing leadership fault line,” Williams-Lindo said. “CIOs are rewarded for minimizing risk; [chief human resources officers] CHROs are rewarded for maximizing capability. AI demands both — and most organizations haven’t reconciled that tension.”
Williams-Lindo described a scenario in which IT focuses on locking down the technical details, while HR is reduced to rolling out “generic training,” leaving employees to translate between the two. Without cooperation across senior leadership teams, silos are likely to form and greatly hinder the employee experience.
Todd Nilson, co-founder of TalentLed Community Consultancy, agreed that AI cannot be left entirely to the CIO to run independently. In fact, he, Williams Lindo and Weed-Schertzer emphasized the importance of not just leveraging IT and HR but also incorporating business line managers across the company, in order to reveal the most meaningful product applications within day-to-day workflows and share those ideas with other functions.
“The most successful implementations I’ve seen are built on cross-functional teams, not owned by one department,” Nilson said.
This doesn’t mean that CIOs have a small role to play; rather, they must cede some ownership over AI if they’re to achieve the returns they want. As Weed-Schertzer put it: “It’s not just a technical product anymore; it’s a reorganization of operations.”
That requires shared leadership and management. It also requires thoughtful employee education.
The difference maker: Training and education
Without sufficient instruction, employees will never be able to get maximum value from AI investment, especially not at scale. Effective training should be tailored to different teams and use cases, but it should also share a common approach: focusing on specific use cases and outcomes, rather than providing granular instruction on which buttons to click.
“If you focus on the tool, it’s going to become procedural,” Weed-Schertzer warned. “‘Here’s how to log in. This is your account.'”
While technically useful, she added that she sees the biggest rewards coming from training employees on specific applications and having managers demonstrate the utility of an AI program for their teams, so that workers have a clear model from which to work. Seeing the utility is what will prompt long-term adoption, as opposed to a demo of basic tool functionality.
CIOs still have a role to play in education. For Williams Lindo, the best training deprioritizes tool expertise in favor of deeper AI literacy. In fact, she argued that effective AI upskilling has almost nothing to do with the tools themselves.
“It’s about judgment,” she said. “People need to know how to interrogate outputs, recognize hallucinations, understand bias and decide when AI should not be used. The companies seeing ROI are building cognitive muscle, not vendor loyalty.”
Nilson supports this emphasis on broader AI understanding as opposed to specific toolkit knowledge. He described AI education as leading employees on a journey, enabling them to visualize how to embed AI into their workflow, rather than simply instructing on functionality. Especially as AI fatigue grows and the shine of these new tools begins to fade, it is critical that management focus on meaningful benefit rather than adoption for adoption’s sake.
“Our job is not merely to inform or even to move them to action,” Nilson said. “It’s to inspire.”
A new, shared path forward
AI is — perhaps uniquely — a technology that employees are likely already exploring in their own time and on personal accounts, developing their own skills and preferences without company oversight.
This puts greater pressure on the CIO to ensure a successful AI rollout. Ignoring employee feedback can be damaging, both by undermining ROI but also by creating security vulnerabilities when a worker uses a preferred but unauthorized AI tool on company devices (known as “shadow AI“). As Nilson explained, it’s human nature to look for the easiest solution — and poor training on authorized tools can easily push employees toward the more well-known, convenient route.
This makes it critical for CIOs to incorporate other stakeholders into the AI implementation process, building in opportunities for feedback from HR, line managers and the users themselves.
“AI success isn’t an IT win; it’s an operating-model shift,” Williams-Lindo said. “CIOs who succeed will stop acting as gatekeepers and start acting as architects of enablement: clear guardrails, shared accountability and trust backed by transparency.”
