The Competition Commission of India (CCI) is expected to release its Market Study on Artificial Intelligence (AI) and Competition this month, which would examine how AI is reshaping markets in India. The study would highlight AI’s pro-competitive potential and its emerging risks, while providing policy recommendations for a competitive ecosystem, sources privy to the matter told Moneycontrol.
The government is also mulling to include AI in the revised Digital Competition Bill (DCB), which is currently being refined to adjust to present market requirements. Moneycontrol reported in August that the DCB might include provisions which allow for a consultative approach while adjudicating cases of competition among tech companies.
AI presents profound challenges to traditional competition regulation, with "algorithmic collusion" emerging as the most significant threat to market competition, say sources. This phenomenon occurs when AI algorithms, operating without explicit human coordination, learn to coordinate pricing or market behaviour independently, creating anti-competitive outcomes that challenge existing legal frameworks, they say.
Therefore, the challenge before CCI is analyse the current regulatory framework, and tweak it to make it capable of handling anti-trust cases, which can penalise "algorithmic collusion" too. Sources say, the CCI is working on enhancing its technical capabilities to probe such cases going forward.
According to experts, India's existing competition framework, anchored by the Competition Act 2002, faces significant limitations in addressing AI-related competition issues. The Act's Section 3, which governs anti-competitive agreements, requires explicit 'agreements' or 'concerted practices' to establish collusion, making it extremely difficult to address autonomous algorithmic behaviour where no human communication or intent exists, they say.
"The traditional requirement of establishing a 'meeting of minds' becomes problematic when algorithms act independently, and proving anti-competitive intent becomes nearly impossible when AI systems operate without human direction. The inherent opacity of many AI systems also makes it difficult for regulators to detect and prove anti-competitive conduct and attribute accountability," explained Akshayy S Nanda, Partner, Saraf and Partners.
Pranjal Prateek, Partner at Khaitan & Co notes that some AI systems can potentially offer superior outcomes in their areas of deployment. For instance, Generative AI search platforms may offer quick, customised and specific results to users which traditional web search may not be able to offer, he says, adding that inputs required for creating the AI architecture could be controlled by a few large players (data, computational power, talent etc), leading to barriers to entry or expansion.
Bharat Budholia, Partner, AZB & Partners says that while the current regulatory framework may be sufficient for AI-related cases, internal training will be critical for the CCI to build technical expertise and understand the complexities of this technology. "There may also be a need to have a more nuanced toolkit to effectively detect and attribute potential competition law violations involving AI," he added.
Other advanced economies, such as US and EU, have formulated frameworks to regulate competition in the emerging AI sector. The EU has adopted a proactive, comprehensive, and centralized regulatory framework, while the US relies on a more fragmented, reactive, and decentralized approach, say experts.
In the US, AI is subject to antitrust scrutiny through the Federal Trade Commission and Department of Justice, which exercises sectoral oversight, without a dedicated AI-specific law. The Federal Trade Commission (FTC) and the DOJ have stated that existing antitrust laws are fully applicable to the AI industry. This includes scrutinizing mergers and acquisitions of AI companies, as well as concerns about "algorithmic collusion".
On the other hand, EU regulates AI-related competition law cases under the EU Competition Laws (Article 101 and 102 of TFEU) and Digital Markets Act (DMA). The DMA's rules on data portability, interoperability, and self-preferencing are designed to prevent large tech companies from leveraging their market power to stifle competition, including in the AI sector.
Discover the latest Business News, Sensex, and Nifty updates. Obtain Personal Finance insights, tax queries, and expert opinions on Moneycontrol or download the Moneycontrol App to stay updated!