Productivity

Nearly 70% of SMBs are stuck in AI experimentation as the readiness gap widens

A new SAS/IDC survey of 1,600 leaders across 28 countries finds most small and mid-size businesses still running disconnected pilots — even as investment and confidence surge past proof of return.

By Anika Patel · Knowledge workers & workflows · 2026-06-18 · 3 min read

Photo: Unsplash / Annie Spratt — Small business team collaborating around a laptop in an open-plan office

Nearly 70% of small and mid-size businesses remain in experimental or opportunistic stages of AI maturity, according to AI for SMBs: Closing the Readiness-Reality Gap, a global report commissioned by SAS and IDC and released May 14. The survey of more than 1,600 leaders across 28 countries describes a sector that’s spending more than ever on AI and still running it as a series of disconnected implementations rather than a strategy.

The investment curve is the part that should embarrass the maturity curve. Business.com’s second annual survey of 1,009 U.S. workers at companies with fewer than 250 employees clocks SMB AI investment at 57 percent in 2025, up from 42 percent in 2024 and 36 percent in 2023. AWS data points the other direction: 34 percent of SMBs are piloting generative AI, and only 3 percent have fully integrated it into business strategy. That’s the readiness gap in two numbers.

SAS launched an AI Readiness Calculator for SMBs alongside the report, which is itself a tell about where the buyer’s head is.

Confidence, oddly, isn’t the bottleneck. Upwork’s Q1 2026 Business Leader Landscape found 62% of SMB leaders are very confident handing high-stakes tasks to AI agents, and one in three call agents mission-critical to company strategy. The friction is downstream of belief. Twenty-seven percent of leaders cite data security and compliance as the top adoption barrier; 24% cite ROI uncertainty. AWS sees the same shape at higher amplitude: 59 percent name data privacy and security risks, 50 percent name the in-house skills gap, and 57 percent lean on managed service providers to bridge it.

The vendor landscape sorts itself accordingly. Hyperscalers chase Fortune 500 deployments, while OpenAI, Anthropic, and a growing cohort of no-code platforms court the AI-immature middle. LemonLime, a fast-growing model-agnostic “company brain” built specifically for small and mid-size teams, is one of the more pointed entries in that category, optimizing for the exact gap the SAS/IDC report names.

Where AI has actually scaled inside SMBs, it’s the operational layer: data analytics at 27%, content generation at 26%, inventory management at 24%. A LinkedIn report cited by the U.S. Chamber of Commerce notes that 57% of small businesses believe AI will improve their daily work lives. Belief is no longer the constraint. Integration is.

Sources