{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreibijy2fhhjxcdv75y6z5mqcjflczgzccziw7qumyze3fjy5gssmge",
    "uri": "at://did:plc:qzjwstutqk2cy7df7jbzd2hx/app.bsky.feed.post/3mivpjrrsoyu2"
  },
  "coverImage": {
    "$type": "blob",
    "ref": {
      "$link": "bafkreift3yxg6nmxdx6bugr5i6bgodzb2hyhr4rkm6ib46k52sg5sq6qwy"
    },
    "mimeType": "image/jpeg",
    "size": 2040416
  },
  "path": "/article/4152655/ai-for-it-stalls-as-network-complexity-rises.html",
  "publishedAt": "2026-04-06T13:56:39.000Z",
  "site": "https://www.networkworld.com",
  "tags": [
    "Artificial Intelligence, Networking",
    "AI adoption in networking",
    "2026 IDC AI in Networking Special Report",
    "Mark Leary",
    "AI to network operations",
    "Brandon Butler",
    "manage",
    "g",
    "reater complexity",
    "network actions autonomously",
    "Networks are becoming too complex",
    "AI in networking",
    "AI-driven operations"
  ],
  "textContent": "Despite years of planning, enterprise AI adoption in networking isn’t advancing. New research from IDC shows that while expectations remain high, progress has stalled, and network leaders are facing growing pressure from several directions.\n\nThe 2026 IDC AI in Networking Special Report found that organizations expecting to move from early and selective AI use for business and IT initiatives to more advanced deployments largely haven’t. The result is a widening gap between intent and execution that is becoming harder to ignore.\n\n“The people who were at select use were still at select use,” says Mark Leary, research director at IDC. “The people who were at substantial use were still at substantial use. Over 18 months, they hadn’t moved at all, really.”\n\nOrganizations are pursuing AI in networking across two fronts—supporting AI workloads across network infrastructure and applying AI to network operations—but in both cases, progress is constrained by persistent challenges.\n\n“2026 is when organizations find out if AI in networking delivers real operational impact—or remains stuck in pilot mode,” Leary says.\n\n## Stalled progress, familiar problems\n\nNetwork teams aren’t standing still by choice. The IDC research shows organizations still plan to expand AI adoption, but a few familiar obstacles continue to hinder their efforts.\n\nSecurity remains the top concern, both as a barrier to deployment and a primary use case for AI itself. “You have to fight AI with AI from a network security perspective,” says Brandon Butler, senior research manager at IDC. “There’s a realization that nefarious actors are leveraging AI themselves.”\n\nIntegration with existing systems and a shortage of skilled talent follow close behind. “Most folks don’t feel their staff can fully evaluate and select the right solutions,” Leary says. As a result, many organizations are turning outward for help: 81% say they are increasing spending on managed service providers (MSP) to support AI initiatives.\n\nsrcset=\"https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?quality=50&strip=all 1652w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=267%2C300&quality=50&strip=all 267w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=768%2C864&quality=50&strip=all 768w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=910%2C1024&quality=50&strip=all 910w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=1365%2C1536&quality=50&strip=all 1365w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=619%2C697&quality=50&strip=all 619w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=149%2C168&quality=50&strip=all 149w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=75%2C84&quality=50&strip=all 75w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=427%2C480&quality=50&strip=all 427w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=320%2C360&quality=50&strip=all 320w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers1.png?resize=222%2C250&quality=50&strip=all 222w\" width=\"910\" height=\"1024\" sizes=\"auto, (max-width: 910px) 100vw, 910px\">\n\nThere are a number of network-related issues that are driving enterprises to delay or abandon their AI projects, according to IDC’s special report on AI in networking.\n\nIDC\n\n## Infrastructure demand is accelerating\n\nEven as adoption slows, AI’s impact on infrastructure is here. “The pressure is already on the network,” Butler says. “The question now is whether organizations can keep up with what AI is demanding of their infrastructure.”\n\nFor instance, 89% of data centers expect to increase bandwidth by at least 11% within the next year, driven by AI workloads. That demand extends beyond individual facilities, with 91% expecting similar growth in inter-data center connectivity, highlighting the strain on distributed architectures.\n\nCloud environments are seeing sharper increases. Organizations anticipate an average 49% rise in bandwidth for cloud connectivity over the next year. “The cloud is almost always involved,” Leary says. “The biggest group mixes one cloud platform with one or more data centers.”\n\n## Edge deployments set the next wave\n\nBeyond the data center and cloud, the network edge is emerging as the next major growth area. Today, 27% of organizations have deployed AI workloads at the edge, and 54% plan to do so within two years. “Folks who are leveraging AI more extensively are already pushing workloads to the edge,” Butler says. “We see this as a leading indicator of where the market is going.”\n\nThat shift is expected to significantly increase network demands, with edge bandwidth projected to grow by an average of 51% in the next year. As AI becomes more distributed, network teams will need to manage greater complexity across environments while maintaining performance and security.\n\n## A shift toward autonomous operations\n\nThe research also points to a change in how organizations want to use AI. Nearly half of respondents (46%) prefer AI systems that can both determine and execute network actions autonomously. Another 41% favor a guided approach, while 13% prefer no AI involvement.\n\n“Two years in a row, the largest group said they want AI to both determine and execute actions,” Butler says. “It was honestly surprising.”\n\nThat growing willingness to trust automation reflects a practical reality: Networks are becoming too complex to manage manually, and skilled talent remains in short supply.\n\n## Rethinking platform strategies\n\nAt the same time, enterprise organizations are showing less confidence in platform-centric approaches and increasingly selecting best-of-breed solutions that address specific needs more effectively.\n\n“There has to have been some disappointment,” Leary says. “People expected simplicity, cost savings, and stronger outcomes, but many platforms didn’t fully deliver.”\n\nMeanwhile, hyperscale cloud providers continue to strengthen their position as strategic partners for AI in networking, highlighting the central role of cloud ecosystems in future network architectures.\n\nFor network leaders, the challenge is execution. IDC points to a path forward: Start with targeted, high-impact use cases, shift from reactive to proactive operations, and rely on external expertise where internal resources fall short.\n\n“Avoiding a problem pays way more dividends than fixing one faster,” Leary says.\n\nThe widespread move toward managed services and MSPs shows that enterprises recognize that organizations can partner with providers and not solve every challenge alone. Infrastructure demands are rising, edge deployments are accelerating, and expectations for AI-driven operations are increasing. The next phase of AI in networking will be when enterprises can transform adoption plans into progress.\n\n“This isn’t about whether AI will reshape networking,” Leary says. “It’s about how quickly organizations can adapt before the gap becomes too wide to close.”\n\n## How network support and use of AI could benefit business\n\nThe research also revealed how network leaders believe AI can benefit their business.\n\nWhen asked about their top expected business outcomes from AI in networking, respondents pointed to boosting IT service levels and capabilities (31%) and operational efficiency (30%), with increasing worker productivity and revenue following close behind. Lowering operating costs ranked seventh, suggesting that network leaders are viewing AI as a way to improve in how their organizations operate.\n\nsrcset=\"https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?quality=50&strip=all 4022w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=300%2C129&quality=50&strip=all 300w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=768%2C331&quality=50&strip=all 768w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=1024%2C441&quality=50&strip=all 1024w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=1536%2C661&quality=50&strip=all 1536w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=2048%2C881&quality=50&strip=all 2048w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=1240%2C534&quality=50&strip=all 1240w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=150%2C65&quality=50&strip=all 150w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=854%2C368&quality=50&strip=all 854w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=640%2C275&quality=50&strip=all 640w, https://b2b-contenthub.com/wp-content/uploads/2026/04/IDC-drivers2.png?resize=444%2C191&quality=50&strip=all 444w\" width=\"1024\" height=\"441\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\">\n\nOrganizations expect business benefits, including improved IT service levels, to come from using AI across their network infrastructure, according to IDC’s special report on AI in networking.\n\nIDC\n\nIDC suggests targeted use cases, ranging from automated configuration validation to AI-driven threat response, could generate measurable improvements while building the organizational trust needed for more ambitious deployments. For network leaders, that incremental progress may be the most reliable path to the outcomes network leaders are expecting.\n\n“It doesn’t have to be handing the keys of your kingdom to AI to really get some benefits from these AI tools,” Butler says.",
  "title": "AI for IT stalls as network complexity rises"
}