Companies across Asia Pacific (Apac) expect artificial intelligence (AI) to deliver strong financial returns in 2026, reinforcing a shift from experimentation toward broader deployment, according to a survey commissioned by Lenovo.
About 88% of organisations in the region expect AI initiatives to generate positive returns this year, with an average projected payoff of 2.8 times the amount invested, or about US$2.85 ($3.67) for every US$1 spent, according to the Lenovo CIO Playbook 2026.
Revenue growth, profitability and improvements in customer and business experience are now the top three priorities for chief information officers, signalling that AI is increasingly tied to commercial performance rather than treated as a technology upgrade.
Yet, scaling remains the central challenge. While expectations for returns are high, many projects struggle to move beyond proof-of-concept stages, underscoring gaps in governance, operating models and lifecycle management.
About 66% of Apac organisations say they are already piloting or systematically deploying AI, while 15% remain in early stages and 19% are still evaluating use cases. In Asean+ markets, 67% report piloting or systematic adoption.
Funding responsibility is also shifting. Half of surveyed organisations say non-IT departments are now paying for AI initiatives, broadening ownership of AI outcomes and elevating the CIO’s role as coordinator across business units rather than sole technology gatekeeper.
See also: Apac businesses feel the squeeze as AI turns customer service into a profit lever: Zendesk
Despite execution hurdles, companies are committing more capital. The majority (96%) of organisations across Apac plan to increase AI investment over the next 12 months, with average budgets expected to rise by about 15%.
Spending will span generative and agentic AI, cloud-based AI services, on-premise infrastructure and security tools. In Asean+, the same proportion of companies plan to lift AI budgets, reinforcing how quickly AI is becoming embedded in enterprise strategy.
“When 96% of organisations are planning a 15% on average increase in AI investment, it tells us that AI decisions are now being made at the core of enterprise strategy. The differentiator will be how effectively organisations integrate AI, embedding it into infrastructure, operations, and security so value compounds over time,” says Sumir Bhatia, President, Asia Pacific, ISG, Lenovo.
See also: AI’s power problem is being overthought, says Microsoft’s VP for energy
“When 96% of organisations are planning a 15% on average increase in AI investment, it tells us that AI decisions are now being made at the core of enterprise strategy,” said Sumir Bhatia, president for Asia Pacific at Lenovo’s Infrastructure Solutions Group. He said the differentiator will be how effectively companies integrate AI into infrastructure, operations and security so that value compounds over time.
Looking ahead, agentic AI is emerging as the next focus area, even as readiness lags ambition. About 21% of Apac organisations report significant use of agentic AI today, while nearly 60% are exploring or planning limited deployments, particularly in telecommunications, healthcare and government, where operational complexity is high.
Only 10% of respondents consider themselves ready for scaled deployment of agentic systems, and 41% say they will need more than 12 months to reach meaningful scale. Security, governance, data quality and integration complexity remain the main barriers.
“Agentic AI represents a fundamental shift in how intelligence is embedded into the enterprise. With nearly 60% of organisations already exploring agentic AI and the majority are choosing a measured path to scale, it reflects that enterprises want AI that operates within core workflows, meets security and governance expectations, and delivers consistent outcomes,” said Fan Ho, executive director and general manager for Asia Pacific at Lenovo’s Solutions and Services Group. He said most companies are taking a measured approach to ensure systems operate within core workflows and meet security and governance requirements before expanding deployments.
As AI workloads scale, infrastructure strategy is emerging as a defining decision for chief information officers. About 86% of Apac organisations now incorporate on-premises or edge environments as part of hybrid AI architectures, effectively making hybrid AI the default model for enterprise deployments. In Asean+, 81% of organisations prefer hybrid architectures, combining on-premise and edge environments to balance performance, security and regulatory requirements.
