The ministry of electronics and information technology in India has issued guidelines with the intent to provide a framework focussing on the balance between innovation in artificial intelligence along with accountability (AI Governance Guidelines). The AI Governance Guidelines will profoundly impact mergers and acquisition (M&A) activities across the deal lifecycle. For corporate development teams, private equity firms, and their advisors, the policy shifts AI from a mere asset to a source of mandatory due diligence and future regulatory risk.
Regulatory compliance and scrutiny
The guidelines introduce new layers of regulatory risk that buyers and sellers must address, particularly when the target company develops or heavily utilizes AI.
-
Due diligence expansion: The scope of M&A due diligence is expanding and there will be an increasing importance in assessing the target company's adherence to forthcoming AI-related regulatory obligations. Buyers must now be vigilant about:
-
Data provenance and ‘dirty data’: Scrutinizing the lawful collection and use of datasets used to train the target's AI models. Issues with ‘dirty data’ or data from problematic sources can significantly impact the deal valuation and may necessitate purchase price adjustments or replacement of faulty algorithms.
Digital Personal Data Protection Act, 2023 (DPDP Act) compliance: The DPDP Act is central to M&A involving data-rich targets. Buyers must verify that the target has obtained explicit consent for processing personal data and has safeguards against misuse, especially when such data is used for AI training.
Algorithm auditability: Buyers are increasingly assessing the target's ability to provide algorithm auditability and transparency especially in light of the anticipated future regulatory obligations under the India AI Governance Guidelines.
Competition and antitrust: The Competition Commission of India (CCI) has also been closely monitoring the intersection of AI and market concentration. Regulators worldwide, and increasingly in India, are concerned about deals that could lead to high levels of concentration or foreclose access to key inputs like proprietary data sets or foundation models. The CCI's ability to review mergers of ‘asset-light digital companies’ based on deal value thresholds (over Rs 20 billion) becomes increasingly relevant for AI startups, where their competitive potential often outweighs current revenue.
Efficiency, predictive analytics, and deal sourcing
For advisors and dealmakers, the guidelines formalize the use of AI tools, driving efficiency across every phase of the deal cycle while emphasizing accountability.
-
Due diligence efficiency: AI technologies, using Natural Language Processing (NLP) and machine learning, are now essential for automating and augmenting traditional due diligence. AI tools can:
-
Accelerate review: Quickly analyze thousands of legal contracts, financial statements, and regulatory filings in a fraction of the time, reducing cost and human error.
Flag hidden risks: Identify anomalies in financial reporting, undisclosed liabilities, and compliance issues that human reviewers might miss.
Predict outcomes: Predictive analytics models can simulate various merger scenarios and outcomes, refining due diligence and integration costs with a high degree of accuracy.
Deal sourcing and valuation: AI eliminates the constraints of traditional deal discovery by continuously scanning public and private data sources (patent filings, hiring trends, social media sentiment) to proactively surface high-fit acquisition targets. This shift is enabling mid-market enterprises and startups to run sophisticated deal-sourcing programs, transforming the episodic, banker-led model into continuous, always-on deal discovery.
AI governance, contracts, and post-merger integration
The ‘sutras of trust and accountability’ mandated by the guidelines necessitate fundamental changes in transactional agreements and post-merger strategies.
-
Contractual protections: Traditional transactional language is often inadequate for AI assets. Purchase agreements must now include tailored clauses that address:
-
IP ownership: Clear ownership and licensing of the AI models, training data, and AI-generated outputs.
Indemnities and escrow: Establishing indemnity provisions and special escrow arrangements to mitigate risks associated with future regulatory non-compliance or the cost of retraining biased models.
Representation on bias: Sellers may be required to affirm that their AI systems are free from safety concerns or discriminatory decisions that could expose the buyer to regulatory scrutiny or consumer claims.
Post-merger integration: AI tools optimize the challenging post-merger integration phase by analyzing workforce synergies, supply chains, and IT systems. Integration efforts must now include:
-
Responsible AI continuity: Ensuring the continuity of responsible AI practices, such as regular performance audits and testing of the acquired systems based on frameworks like ISO/IEC 42001:2023.
Integration of governance: The buyer must integrate the target's AI governance into their own framework, aligning for a ‘whole-of-government’ regulatory approach that affects all sectors under the AI Governance Guidelines.
The AI Governance Guidelines ensure that M&A in India remains a strategic avenue for growth. However, success now hinges on performing AI-specific due diligence that views the technology not just as an asset, but as a potential liability requiring deep and verifiable compliance.
Sameer Sah is a Partner and Rakshitha Naik is a Senior Associate with Khaitan & Co. Views are personal.
