Urgent.News

What's breaking now, across thousands of outlets.

AI

FBR introduces rules for electronic scrutiny of taxpayers

ISLAMABAD: The Federal Board of Revenue (FBR) has begun extending the use of artificial intelligence to scrutinise individual tax returns, enabling its automated system to detect factual and legal mistakes and discrepancies before initiating legal or penal action. The draft rules, titled “38-B Procedure for electronic scrutiny and intimation of issues detected by the automated system”, have been…

FBR introduces rules for electronic scrutiny of taxpayers

Islamabad: The Federal Board of Revenue (FBR) is rolling out new rules to use artificial intelligence for checking individual tax returns. The automated system will spot factual and legal errors before taking legal or penal action. The draft rules, titled "38-B Procedure for electronic scrutiny and intimation of issues detected by the automated system", are being proposed through amendments to the Income Tax Rules 2002.

After receiving feedback from stakeholders, the final rules will be set within three days. A major aspect of the plan is the automated comparison of data from tax authorities and income tax returns to spot potential mistakes. The system will alert taxpayers via the IRIS portal about any discrepancies, giving them a chance to correct errors before legal action is taken.

Taxpayers will receive electronic notice of the identified issues and have at least seven days to clarify or fix them. If a taxpayer doesn't respond within the time frame, they will be given another reminder. The Inland Revenue officer will then review the taxpayer's response before deciding on further action according to the Income Tax Ordinance and rules.

Written by urgent.news from Dawn - Pakistan's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Also reported by 1 other outlet

Read the original at dawn.com →

More in AI

Day 6: Data Preprocessing — Cleaning the Messy Reality of Enterprise Data

Real-world enterprise data is rarely ready for machine learning. Whether you are analyzing console output from high-performance networking hardware, such as troubleshooting transceiver EEPROM data on…

  • Raw enterprise data is messy and requires cleaning before effective use
  • Steps include handling missing values, managing outliers, and deduplication
  • Proper preprocessing turns chaotic data into clean format for ML algorithms

More from Wednesday 7 October →