﻿<?xml version="1.0" encoding="utf-8"?>
<!--<!DOCTYPE nitf SYSTEM "nitf-3-4.dtd">-->
<nitf>
  <head>
    <title id="Title">#Title</title>
    <docdata management-doc-idref="">
      <date.issue id="CreationDate" norm="" />
      <du-key id="rev-ver" generation="1" version="Default" />
      <du-key id="Parent-Version" version="" />
      <identified-content>
        <classifier id="newspro-nitf" value="r2" />
        <classifier id="Newspro-App" value="Epaper" />
        <classifier id="Content-Type" value="Story" />
        <classifier id="storyID" value="" />
        <classifier id="CmsConID" value="" />
        <classifier id="Desk" value="" />
        <classifier id="Source" value="" />
        <classifier id="Edition" value="" />
        <classifier id="Category" value="-1" />
        <classifier id="UserName" value="" />
        <classifier id="PublicationDate" value="22/08/2026" />
        <classifier id="PublicationName" value="HI" />
        <classifier id="IsPublished" value="Y" />
        <classifier id="IsPlaced" value="Y" />
        <classifier id="IsCompleated" value="N" />
        <classifier id="IsProofed" value="N" />
        <classifier id="User" value="" />
        <classifier id="Headline-Count" value="" />
        <classifier id="Slug-Count" value="0" />
        <classifier id="Photo-Count" value="0" />
        <classifier id="Caption-Count" value="0" />
        <classifier id="Word-Count" value="0" />
        <classifier id="Character-Count" value="0" />
        <classifier id="Location" value="" />
        <classifier id="TemplateType" value="1" />
        <classifier id="StoryType" value="Story" />
        <classifier id="Author" value="" />
        <classifier id="UOM" value="mm" />
        <classifier id="NumCol" value="0" />
        <classifier id="kicker" value="" />
        <classifier id="ByLine" value="" />
        <classifier id="DateLine" value="" />
        <classifier id="box-geometry" value="36,1171,488,1485" />
        <classifier id="Layer" value="Default" />
        <classifier id="numcol" value="4" />
        <classifier id="ArticleStyle" value="" />
        <classifier id="Epaper-Build" value="7.96.0.0" />
        <classifier id="ProcessingDateTime" value="Fri Aug 21 2026 23:26:36 GMT+0530" />
      </identified-content>
      <urgency id="home-page" ed-urg="0" />
      <urgency id="priority" ed-urg="0" />
      <doc-scope id="scope" value="0" />
    </docdata>
    <pubdata type="print" name="HI" date.publication="20260822T000000+5.30" edition.name="VJA" edition.area="VJA" position.section="22MAIN07FVJA" position.sequence="7" ex-ref="22MAIN07FVJA.indd" />
  </head>
  <body boxBorderWeightColor="" boxBorderWeight="">
    <body.head>
      <hedline>
        <hl1 id="Headline1" class="1" style="Headline1">
          <lang class="3" style="Headline1" font="Franklin Gothic Demi Cond" fontStyle="Regular" size="13">National GST meeting</lang>
        </hl1>
        <hl2 id="Headline1" class="1" style="Headline2">
          <lang class="3" style="Headline2" font="Chronicle Display" fontStyle="Italic" size="25">AP govt’s AI-powered GST model draws wide attention</lang>
        </hl2>
        <hl3 id="Headline1" class="1" style="Headline3">
          <lang class="3" style="Headline3" font="Franklin Gothic Demi Cond" fontStyle="Regular" size="16">Legal-AI Officer Asst trained on GST laws, 22,000 judgments</lang>
        </hl3>
      </hedline>
    </body.head>
    <body.content id="Bodytext" CaptionAsBody="0">
      <block>
        <media id="1" media-type="image">
          <media-reference id="tn" source-credit="" data-location="1" source="ChatGPTImageA_7_VJA_tn.jpg" Units="pixels" width="50" height="50"></media-reference>
        </media>
      </block>
      <p style=".Bodylaser">
        <lang class="3" style=".Bodylaser" font="Minion Pro" fontStyle="Regular" size="9">Hans News Service
Amaravati</lang>
      </p>
      <p style=".Bodylaser">
        <lang class="3" style=".Bodylaser" font="Minion Pro" fontStyle="Regular" size="9">AndhraPradesh has successfully showcased its pioneering GST administration model, based on artificial intelligence (AI) and machine learning (ML), at the sixth National Coordination Meeting held at Vigyan Bhawan in New Delhi on Friday. The tech-driven model, appreciated widely, now has potential to emerge as a benchmark and replicable model for GST administration across the country.</lang>
      </p>
      <p style=".Bodylaser">
        <lang class="3" style=".Bodylaser" font="Minion Pro" fontStyle="Regular" size="9">The meeting, chaired by the Union revenue secretary, was attended by the GST Council Secretariat, CBIC and senior tax officials from States and Union Territories.</lang>
      </p>
      <p style=".Bodylaser">
        <lang class="3" style=".Bodylaser" font="Minion Pro" fontStyle="Regular" size="9">Andhra Pradesh’s presentation, titled “Use of AI/ML and Automation in GST Tax Administration — From Case Selection to Litigation,” was included as a dedicated agenda item. The Commercial Taxes Department explained how AI-based systems are being used for case selection and allocation, return scrutiny under Section 61, audit under Section 65, inspection under Section 67 and litigation before the First Appellate Authority, GST Appellate Tribunal and High Court. The system integrates data from four sources and uses 19 analytical reports and a 35-parameter risk matrix to identify and classify cases before they reach officers. High-value cases are routed for additional scrutiny through statutory mechanisms. The department also demonstrated its Legal-AI Officer Assistant, trained on GST laws and about 22,000 judgments from courts and tribunals. It has been extended across three litigation forums and is assisting officers with drafting instructions, para-wise remarks, and replies for more than 13,700 cases.</lang>
      </p>
      <p style=".Bodylaser">
        <lang class="3" style=".Bodylaser" font="Minion Pro" fontStyle="Regular" size="9">The department presented data comparing AI-assisted operations with the earlier manual system. During six months of AI-driven return scrutiny, revenue detection reached Rs 743.43 crore, compared with Rs 365.75 crore during the comparable legacy period. Detection per case also rose sharply to Rs 27.63 lakh, against Rs 3.08 lakh under the earlier process.</lang>
      </p>
      <p style=".Bodylaser">
        <lang class="3" style=".Bodylaser" font="Minion Pro" fontStyle="Regular" size="9">The department stressed that AI does not replace statutory decision-making. It generates drafts and analytical inputs, while officers take decisions and digitally sign notices and orders, ensuring traceability.</lang>
      </p>
      <p style=".Bodylaser">
        <lang class="3" style=".Bodylaser" font="Minion Pro" fontStyle="Regular" size="9">The presentation received strong interest from participating States. Tamil Nadu, Bihar, Rajasthan, Chhattisgarh, Kerala, Assam and Meghalaya have sought details of the system and are planning visits to Andhra Pradesh to study its architecture and workflow.</lang>
      </p>
      <p style=".Bodylaser">
        <lang class="3" style=".Bodylaser" font="Minion Pro" fontStyle="Regular" size="9">The Revenue Secretary commended AP’s initiative as a model that other States could consider adopting.</lang>
      </p>
    </body.content>
  </body>
</nitf>