Research Articles Open Access

Governance of Generative AI in Bank Marketing Content in Pakistan: Copyright, Trade Mark Protection, Deepfake Impersonation, and Compliance by Design

Madiha Awais
Available online July 1, 2025
PDF

Abstract

Generative artificial intelligence (GenAI) can be used in bank marketing to prepare advertising copy, fake images, conversational content, and personalized offers. Its use in banking also creates problems that are not limited to marketing quality. Incorrect product information, unauthorized use of protected material, fake endorsements, or impersonated content can affect customers and expose banks to legal and regulatory concerns. In this paper, we propose a governance and compliance framework for the use of GenAI in bank marketing in Pakistan. The framework covers rights clearance for source material and generated content, review of product claims, disclosure and content provenance, privacy and fairness in personalized marketing, and response to deepfake-enabled impersonation and brand spoofing. We develop these controls with reference to Pakistan's laws and regulatory requirements concerning trademarks, copyright, deceptive marketing, consumer protection, customer data, and cybercrime. Relevant banking technology-governance requirements and international AI-governance principles are also considered. A scenario-based case study is used to explain how the controls apply during content preparation, review, publication, targeting, and post-publication monitoring. The framework also defines responsibilities and records that can be used for review and audit. Its effectiveness is not empirically evaluated in this study and remains a subject for future work.

Keywords

Generative AI Banking Marketing Copyright Trade Marks Deepfakes

Introduction

Banks in Pakistan communicate with customers through several digital channels, including mobile applications, digital banking platforms, social-media campaigns, personalized offers, chatbots, and other conversational interfaces. Customers utilize these messages to learn about financial products, fees, eligibility requirements, and available services, thus their reliability is crucial. It is now possible to create advertising copy, customer messaging, product descriptions, campaign images, and various forms of promotional content using GenAI, which includes massive language models and image-generation systems. An activity that already relies on precise information and unambiguous identification of the bank as the source of communication is made riskier by its usage. Another challenge is use of deceptive or protected content. A produced phrase, picture, logo, or campaign design can be similar to another brand or imply an unapproved affiliation. When copyrighted content is fed into a GenAI system or when the output replicates protected expression, copyright problems may also arise. Another issue is that deepfake movies, cloned sounds, and synthetic people might lead to customers accepting altered content as official bank correspondence. External actors can also utilize GenAI to replicate a bank’s identity and spread phony promotional messages or ads. As a result, the issue is not exclusive to information produced by banks. It also covers the abuse of the bank’s brand by using fake and fake material.

Parts of this issue have been addressed in earlier research. Duivenvoorde [1] examines GenAI-based marketing from the standpoint of consumer protection and talks about transparency, accountability, and enforcement. In [2], the authors investigate how consumers evaluate the legitimacy and authenticity of synthetic adverts in order to determine the impact of transparency in deepfake advertising. Concerns about copyright and other intellectual property have also been mentioned as obstacles to using GenAI in marketing [3]. These studies are helpful for comprehending specific concerns, but they don’t explain how these risks might be managed collectively in Pakistan’s financial and judicial systems.

The usage of GenAI in bank marketing within Pakistan’s current institutional and regulatory framework is examined in this article. In addition to consumer protection and competition law concepts pertaining to deceptive advertising and unfair market practices [4], the study takes into account the Trade Marks Ordinance 2001 and the Copyright Ordinance 1962. Impersonation, fraudulent campaigns, and misuse of synthetic media are also taken into consideration under the Prevention of Electronic Crimes Amendment Act 2025 [5]. Based on these requirements, we propose a governance approach for GenAI-assisted bank marketing and relate it to the State Bank of Pakistan’s technology-governance expectations and relevant international AI governance standards [6,7]. The study is limited to bank marketing and related content operations. The research question addressed in this paper is: how can banks in Pakistan use GenAI for marketing and content operations while maintaining legal accountability, protecting customers, preserving brand integrity, and safeguarding fundamental rights? The main contributions of this paper are given below:

  • We identify the main legal and governance risks that arise when GenAI is used in bank marketing in Pakistan and relate these risks to the relevant legal obligations and enforcement considerations.

  • We propose a compliance framework in which intellectual property, consumer protection, transparency, privacy, fairness, and deepfake response are handled within the bank marketing process.

  • Two workflow figures are used to present the governance controls for GenAI-assisted marketing and the response process for deepfake-enabled bank impersonation.

  • The practical implementation of the proposed restrictions is demonstrated through a case study of the banking industry. Additionally, the study identifies indicators that may be utilized for empirical evaluation and auditing in the future.

The remainder of the paper is organized as follows. In Section 2, related work on deepfake disclosure, consumer protection, GenAI marketing, intellectual property, and governance requirements is presented. The legal and regulatory environment of Pakistan is covered in Section 3. The study’s scope and research design are described in Section 4. The suggested compliance structure is presented in Section 5, and the deepfake and impersonation response sequence is explained in Section 6. The case study of the banking industry and the suggested audit indicators are presented in Section 7. The results and restrictions are described in Section 8. The paper is concluded in Section 9.

Complete Article

The complete article, including all figures, tables, equations and algorithms, is available in the official publication PDF.

Conclusion

Although GenAI may be used for content creation, customization, and client engagement in bank marketing, its application also brings up concerns about copyright, trademarks, false claims, customer data, synthetic media, and bank impersonation. These concerns must be taken into account in Pakistan together with consumer and competition legislation, cybercrime regulation, data protection policy, financial technology governance, and intellectual property law. We presented a compliance framework for GenAI-assisted bank marketing in Pakistan in this article. At various phases of the marketing process, the framework places rights clearance, brand review, product claim verification, customer data controls, disclosure, content provenance, and deepfake reaction. The use of these controls was shown through a scenario-based banking case study, and a set of indicators was established for further audit and assessment. The paper does not demonstrate that putting the framework into practice will lessen regulatory risk, fraud, or litigation. Such impacts need to be evaluated empirically. The suggested approach gives banks an organized method for allocating accountability, documenting significant choices, and implementing the necessary controls during GenAI-assisted marketing.

References

  1. Duivenvoorde, B. (2025). Generative AI and the future of marketing: A consumer protection perspective. Computer Law & Security Review, 57, 106141. https://doi.org/doi: 10.1016/j.clsr.2025.106141
  2. Whittaker, L., Mulcahy, R., Russell-Bennett, R., Letheren, K., & Kietzmann, J. (2025). Examining consumer appraisals of deepfake advertising and disclosure: Show deepfakes as “real life” or say they're “just fantasy”? Journal of Advertising Research, 66(1), 113–134. https://doi.org/doi: 10.1080/00218499.2025.2498830
  3. Sutrisno, G. B. (2025, January 30). WFA finds copyright and IP issues are stalling GenAI in marketing. Marketing-Interactive. https://www.marketing-interactive.com/wfa-finds-copyright-and-ip-issues-are-stalling-genai-in-marketing
  4. Government of Pakistan, Ministry of Law and Justice. (2001). Trade Marks Ordinance, 2001 (Ordinance No. XIX of 2001). Pakistan Code. https://pakistancode.gov.pk/
  5. Government of Pakistan. (2025). Prevention of Electronic Crimes (Amendment) Act, 2025 (Act No. II of 2025). The Gazette of Pakistan, Extraordinary, January 29, 2025. https://na.gov.pk/uploads/documents/679b243193585_457.pdf
  6. State Bank of Pakistan. (2017). Enterprise Technology Governance & Risk Management Framework for Financial Institutions. Annex to BPRD Circular No. 05 of 2017, Banking Policy & Regulations Department. https://www.sbp.org.pk/bprd/2017/C5-Annex.pdf
  7. Tabassi, E. (2023). Artificial Intelligence Risk Management Framework (AI RMF 1.0). NIST AI 100-1. National Institute of Standards and Technology. https://doi.org/doi: 10.6028/NIST.AI.100-1
  8. Chesterman, S. (2025). Good models borrow, great models steal: Intellectual property rights and generative AI. Policy and Society, 44(1), 23–37. https://doi.org/doi: 10.1093/polsoc/puae006
  9. Budd, F., Gray, L., & Aries, N. (2025). The impact of AI on trade mark law and practice. Trade Marks Laws and Regulations Report 2025, International Comparative Legal Guides (ICLG), Global Legal Group. https://iclg.com/practice-areas/trade-marks-laws-and-regulations/01-the-impact-of-ai-on-trade-mark-law-and-practice
  10. International Organization for Standardization & International Electrotechnical Commission. (2023). ISO/IEC 42001:2023: Information technology–Artificial intelligence–Management system. ISO. https://www.iso.org/standard/42001.html
  11. Government of Pakistan, Ministry of Law and Justice. (1962). Copyright Ordinance, 1962 (Ordinance No. XXXIV of 1962). Pakistan Code. https://pakistancode.gov.pk/pdffiles/administrator5d255d1f44b50b65d6a3e87a2e0d3fee.pdf
  12. Government of Pakistan. (2010). Competition Act, 2010 (Act No. XIX of 2010). Pakistan Code. https://pakistancode.gov.pk/pdffiles/administratore1b5ab79c2c1975670e6aa7fe57e57d6.pdf
  13. Government of Punjab. (2005). The Punjab Consumer Protection Act 2005 (Pb. Act II of 2005). https://pccmdpunjab.gov.pk/custom/mateen/pdf/ruleregulation/20-Punjab%20Consumer%20Protection%20Act%202005.pdf
  14. Ministry of Information Technology and Telecommunication, Government of Pakistan. (2023). Personal Data Protection Bill, May 2023 (Final Draft). https://www.moitt.gov.pk/Legislations
  15. State Bank of Pakistan. (2017). BPRD Circular No. 05 of 2017: Enterprise Technology Governance & Risk Management Framework for Financial Institutions. Banking Policy & Regulations Department, May 30, 2017. https://www.sbp.org.pk/circulars/bprd-circular-no-05-of-2017
  16. State Bank of Pakistan. (2019). BPRD Circular No. 06 of 2019: Framework for Risk Management in Outsourcing Arrangements by Financial Institutions. Banking Policy & Regulations Department, December 17, 2019. https://www.sbp.org.pk/circulars/bprd-circular-no-06-of-2019
  17. Coalition for Content Provenance and Authenticity. (2025). Content Credentials: C2PA Technical Specification, Version 2.4. https://spec.c2pa.org/specifications/specifications/2.4/specs/C2PA_Specification.html
  18. Romero-Moreno, F. (2025). Deepfake detection in generative AI: A legal framework proposal to protect human rights. Computer Law & Security Review, 58, 106162. https://doi.org/doi: 10.1016/j.clsr.2025.106162
34 14

Similar Articles

You may also start an advanced similarity search for this article.