In a major move to safeguard national interest and sensitive financial data, the Union Ministry of Finance has issued a strict directive to all its employees prohibiting the use of generative AI platforms, specifically naming ChatGPT and the Chinese-origin DeepSeek, for official work.
The advisory marks a significant step in the Indian government’s cautious approach toward the rapid integration of external artificial intelligence tools in administrative processes.
The Core Concerns: Privacy and Sovereignty
The ministry’s decision is rooted in several critical security risks associated with public AI models:
-
Data Leakage: There is a high risk that employees might inadvertently upload confidential government documents, policy drafts, or internal memos into AI prompts to summarize or rephrase them.
-
Third-Party Storage: Since these AI models process data on external, often foreign-based servers, the government loses control over how sensitive information is stored, used, or potentially accessed by third parties.
-
Foreign Origins: The specific mention of DeepSeek highlights growing concerns regarding data residency and the potential for foreign surveillance or influence through AI platforms developed outside India.
Key Directives for Employees
The official communication outlines clear boundaries for ministry staff:
-
Strict Prohibition: No official notes, cabinet papers, or classified data are to be fed into any public AI chatbot.
-
Use of Authorized Tools: Employees are encouraged to rely on internal government-approved systems and the National Informatics Centre (NIC) infrastructure for digital tasks.
-
Accountability: Any breach of these guidelines that results in the exposure of sensitive government information will be treated as a serious disciplinary matter.
A Growing Global Trend
The Finance Ministry’s move mirrors actions taken by several global corporations and governments. From Wall Street banks to international tech giants, organizations are increasingly blocking public AI tools to prevent “shadow AI” (the unauthorized use of AI by employees), which can lead to intellectual property theft or regulatory non-compliance.
The Road Ahead: Indigenous AI?
While the ban limits the use of public tools, it paves the way for the development of “Sovereign AI.” The Indian government is reportedly exploring the creation of its own large language models (LLMs) that are hosted locally and trained on secure datasets, ensuring that the benefits of AI productivity are realized without compromising national security.
The Bottom Line: For the Finance Ministry, the efficiency gains offered by AI do not outweigh the potential cost of a data breach. The directive reinforces a “security-first” culture within the nation’s highest financial decision-making body.
