2026 APAC AI Sentiment Survey Report: adoption is near universal; governance is not

New LexisNexis® research among 1,715 legal professionals in 10 Asia-Pacific markets examines how AI is creating value, what legal professionals need to trust it, and how they want to use it next.

This survey finds that 94% of respondents use AI in their work, up from 88% in 2025. Yet only 58% work under an AI policy, and just 54% say their organisation assesses the value AI delivers.

Across APAC legal markets, AI is already embedded in everyday legal work. The more revealing differences now lie in the structures surrounding that use: whether organisations have established clear policies, whether they can demonstrate value, and whether AI fits the progression of a legal task.

Taken together, the findings describe a legal profession that has embraced AI faster than it has developed the governance, measurement, and workflows needed to make its value dependable.

Download the full report →

APAC legal AI in 2026: the findings at a glance

• AI use is widespread: 94% use general-purpose AI, specialist legal AI, or both
• Legal professionals expect greater capacity: 83% agree AI can create more capacity for higher-value work
• Governance is still developing: 58% report that their organisation has an AI policy
• Value measurement remains uneven: 53% say their organisation assesses the value of AI use
• Greater structure could encourage further use: 69% would use AI more if it were delivered through guided workflows
• Trust is the constraint: 63% cite accuracy and hallucination risk as their leading concern

Adoption is widespread, but it is not the whole story

Almost half of respondents now use specialist legal AI, either on its own or alongside general-purpose tools (46%). Its use is concentrated in legal research, drafting, document review, and legal analytics, where authority, legal context, and the ability to verify an answer carry particular weight.

The distinction matters because widespread use says little about the quality of output, the value created, or how well the technology fits the wider legal workflow.

Value is visible, but rarely measured

The clearest expected benefit is more capacity for work requiring legal judgment and expertise. Most respondents agree that reducing lower-value work can create greater capacity for higher-value activity (83%).

Measurement, however, remains inconsistent. Just over half say their organisation assesses the value of AI use (53%), most commonly by looking at time savings and output accuracy.

Time saved at one stage of a task may be absorbed later through verification, correction, or transferring work between disconnected systems. Understanding value therefore, requires organisations to consider the complete legal task, rather than the speed of an initial output alone.

Trust and workflow will shape what comes next

Accuracy and hallucination risk remain the leading concerns among respondents (63%), followed by confidentiality and data security (57%). These concerns do not measure how often errors or breaches occur, but they help explain why trust remains integral to the practical value of AI.

Legal professionals also want more structure. More than two-thirds say they would be more likely to use AI if it were delivered through guided workflows (69%).

That preference points towards AI that supports the progression of a legal task, with opportunities for the user to review and direct the approach. For legal teams, the next phase will depend on connecting AI with authoritative information, professional oversight, and workflows that reflect how legal work is carried out.

Methodology: LexisNexis surveyed 1,715 legal professionals across Australia, the Chinese Mainland, Hong Kong SAR, India, Indonesia, Japan, Korea, Malaysia, New Zealand, and Singapore between 22 May 2026 and 29 June 2026. Participants were recruited from LexisNexis customers and prospective customers.

Complete the form to download the full 2026 APAC AI Sentiment Survey report.

.