Vijayawada: Andhra Pradesh’s artificial intelligence (AI)-driven GST administration model has received appreciation from the union revenue secretary and several states at the national GST coordination meeting held at Vigyan Bhawan in New Delhi on Friday.The model, presented by Babu Ahmed, chief commissioner of state taxes, AP, focused on the use of AI and machine learning to improve tax compliance, taxpayer services and enforcement. The meeting was attended by senior officials from the GST council secretariat, CBIC and commercial tax departments of states and Union territories.“AP has demonstrated that AI can be effectively integrated into the statutory tax administration process without compromising the role of officers. The technology assists in identifying cases and preparing drafts, while the final decision remains with the authorised officer,” Babu Ahmed said.The presentation, titled “Use of AI/ML and Automation in GST Tax Administration — From Case Selection to Litigation”, covered the entire tax administration cycle, including case selection, return scrutiny, audit, inspection and litigation.The department has developed a data integration system drawing information from four sources and uses 19 automated analytical reports and a 35-parameter risk matrix to identify and classify cases before they are assigned to officers. Value-based statutory routing ensures that high-value cases receive additional scrutiny at senior levels.“We have also showcased its legal-AI officer assistant, which has been trained on GST laws and around 22,000 judgments of the Supreme Court, high courts, tribunals and first appellate authority. The system has been deployed across three litigation forums and has been used in more than 13,700 cases to assist officers in preparing instructions, parawise remarks and replies,“ said Babu Ahmed.The presentation drew interest from several states, including Tamil Nadu, Bihar, Rajasthan, Chhattisgarh, Kerala, Assam and Meghalaya, which sought details of AP’s architecture and workflow. Teams from some of these states are expected to visit AP to study the system for possible replication. Babu said that they were planning to extend AI-based risk scoring to provisional refunds, GST registration monitoring and cancellation, revision and review petitions, arrears and recovery proceedings, and advance rulings. AP has also sought greater cooperation from GSTN, including direct API integration, wider taxpayer data visibility and reliable access to APIs during peak periods, to further scale its AI-based tax administration system.GFX– The AI-assisted return scrutiny had produced significant results– The revenue detection stood at Rs 743.43 crore during a six-month period, compared with Rs 365.75 crore during the corresponding legacy period– Detection per case increased to Rs 27.63 lakh from Rs 3.08 lakh.– The department had ensured that automation remained within the statutory framework.–AI drafts, but officers decide. Every notice and order carries the digital signature of the responsible officer and is fully traceable through the back office case ID and document identification number ~Babu Ahmed, chief commissioner, state taxes, AP
Leave a Reply