Despite extensive planning efforts, the adoption of artificial intelligence (AI) within enterprise networking is faltering. Recent research reveals a stark contrast between the ambitions of organizations and the realities they face in executing these plans, particularly as network complexities continue to grow.
Enterprise AI Adoption Stalls
The 2026 IDC AI in Networking Special Report highlights that while many organizations initially aimed to transition from limited AI applications to more advanced implementations, progress has largely stagnated. Mark Leary, research director at IDC, notes, “The people who were at select use were still at select use. The people who were at substantial use were still at substantial use. Over 18 months, they hadn’t moved at all, really.”
This stagnation signifies a widening gap between organizations' intentions regarding AI and their actual execution, which is increasingly difficult to overlook.
Ongoing Challenges in AI Integration
Organizations are attempting to leverage AI across two main fronts: supporting AI workloads across network infrastructure and utilizing AI to enhance network operations. However, persistent challenges continue to impede their efforts.
“2026 is when organizations find out if AI in networking delivers real operational impact—or remains stuck in pilot mode,” says Leary.
Familiar Obstacles
Security concerns remain a primary barrier to AI deployment, as well as a significant use case for AI technologies. Brandon Butler, senior research manager at IDC, states, “You have to fight AI with AI from a network security perspective. There’s a realization that nefarious actors are leveraging AI themselves.”
Integration issues with existing systems and a shortage of skilled talent are also major hurdles. Many organizations report feeling unprepared to evaluate and select the right AI solutions, prompting 81% of them to increase spending on managed service providers (MSPs) to support their AI initiatives.
Rising Infrastructure Demands
As adoption rates stall, the demands placed on network infrastructure by AI continue to escalate. Butler emphasizes, “The pressure is already on the network. The question now is whether organizations can keep up with what AI is demanding of their infrastructure.”
A notable 89% of data centers anticipate increasing bandwidth by at least 11% within the next year, driven by the growth of AI workloads. This demand extends to inter-data center connectivity, with 91% expecting similar growth, underscoring the strain on distributed architectures.
Cloud environments are experiencing even sharper increases, with organizations predicting an average 49% rise in bandwidth for cloud connectivity in the coming year.
Edge Deployments as a Growth Area
Beyond traditional data centers and cloud services, edge computing is emerging as a significant growth area for AI deployment. Currently, 27% of organizations have initiated AI workloads at the edge, and 54% plan to do so within the next two years. Butler notes, “Folks who are leveraging AI more extensively are already pushing workloads to the edge.”
This shift is expected to dramatically increase network demands, with edge bandwidth projected to grow by an average of 51% in the next year.
Autonomous Operations on the Horizon
The research indicates a shift in organizations' preferences regarding AI deployment. Nearly half of the respondents (46%) prefer AI systems capable of autonomously determining and executing network actions. This reflects a growing willingness to embrace automation as networks become increasingly complex and skilled personnel remain scarce.
Rethinking Platform Strategies
Enterprise organizations are reassessing their confidence in platform-centric approaches and are increasingly opting for best-of-breed solutions tailored to their specific needs. Leary mentions, “There has to have been some disappointment. People expected simplicity, cost savings, and stronger outcomes, but many platforms didn’t fully deliver.”
Hyperscale cloud providers are strengthening their positions as strategic partners for AI in networking, emphasizing the essential role of cloud ecosystems in future network architectures.
Path Forward for Network Leaders
For network leaders, the focus now shifts to execution. IDC suggests starting with targeted, high-impact use cases, transitioning from reactive to proactive operations, and leveraging external expertise where internal resources are limited. Leary emphasizes, “Avoiding a problem pays way more dividends than fixing one faster.”
The trend towards managed services indicates that enterprises are recognizing the value of partnerships in overcoming challenges. With rising infrastructure demands, accelerating edge deployments, and increasing expectations for AI-driven operations, the next phase of AI in networking hinges on successfully transforming adoption plans into tangible progress.
“This isn’t about whether AI will reshape networking,” Leary concludes. “It’s about how quickly organizations can adapt before the gap becomes too wide to close.”
Expected Business Benefits from AI in Networking
The research also reveals that network leaders anticipate several business benefits from AI integration. The most significant expected outcomes include enhanced IT service levels (31%) and operational efficiency (30%), with improved worker productivity and revenue also ranking high. Interestingly, lowering operating costs is seen as less of a priority, indicating a perspective that views AI primarily as a means to enhance operational capabilities.
Targeted use cases, ranging from automated configuration validation to AI-driven threat responses, are projected to yield measurable improvements while simultaneously building the organizational trust necessary for broader AI deployment.
Butler concludes, “It doesn’t have to be handing the keys of your kingdom to AI to really get some benefits from these AI tools.”
Source: Network World News