Zifo has been named runner-up in the 2026 CDISC AI Innovation Challenge for TLFGenix, its artificial intelligence-powered system designed to automate the generation of tables, figures and listings used in clinical trial reporting.
The recognition came under the challenge’s third use case, AI-Driven Tables, Figures, and Listings Generation. TLFGenix is designed as an end-to-end AI pipeline that takes Statistical Analysis Plans, or SAPs, and Analysis Data Model datasets as inputs and converts them into traceable clinical reporting outputs.
According to Zifo, the system uses six specialised AI agents to handle different stages of the reporting process. These agents digitise the SAP into a structured model, profile ADaM data, generate an analysis specification aligned with CDISC's Analysis Results Standard, produce R driver scripts and execute them in a network-isolated environment. The system also checks schemas and output consistency.
Human oversight remains part of the workflow. Statisticians approve the analysis specification before code generation and sign off on individual tables, figures and listings. Zifo said its models operate on internal infrastructure, allowing clinical data to remain within the enterprise environment.
TLFGenix also creates a lineage graph connecting study objectives, endpoints, analyses and results. If an approved specification or SAP changes, the system can identify affected outputs for selective regeneration and further approval. Each output is accompanied by an evidence record containing information such as generated code, execution logs, approvals and a traceability matrix.
The development targets an area of clinical research that traditionally involves specialised statistical programming and extensive review. Zifo positions the system as a way to reduce repetitive programming while retaining human review, traceability and auditability rather than replacing statistical judgement.
The 2026 CDISC AI Innovation Challenge received 50 submissions from 30 organisations and individual entrants, compared with 22 submissions in its inaugural 2025 edition. The competition explores how artificial intelligence and machine learning can support the adoption and application of clinical research data standards.
This marks Zifo’s second consecutive year receiving recognition in the challenge. In 2025, the company was named runner-up in two categories involving protocol digitisation and automated semantic traceability.
TLFGenix is scheduled to be presented alongside other winning and runner-up solutions at the 2026 CDISC US Interchange in Denver in October.
Disclaimer: This article may include information derived from interviews, press releases, public statements, research, company communications and other publicly available or third-party sources. Such material may be summarised, paraphrased or contextualised for journalistic and editorial purposes. All rights in third-party content remain with their respective owners.