Is an advisory enough to tackle risks flagged by the Competition Commission of India?
India’s competition regulator warns AI could warp markets against the consumer’s interests. Yet its advice stops at self-regulation. With AI adoption all the rage, risks range from AI-erected market barriers to algorithmic collusion. Can moral suasion stop AI deployment going down a slippery slope?
There is an undefined organism in town called artificial intelligence (AI). It is undefined because nobody knows what eventual form or shape it will acquire; its nascent presence has already evoked a mix of emotions: fear, despair and hope. It is an organism because it’s growing at an alarming pace in its dimensions as well as in its faux sentience.
Businesses are rushing to adopt AI in the hope that its embrace will help them improve efficiency, upgrade productivity and strengthen their bond with customers. Some Indian businesses, loath to spend time and money on delving deep, are rushing to bolt AI onto existing operations without an adequate grasp of its scope, hoping that this will miraculously deliver results.
Many display ignorance of collateral damage. Others are aware of AI risks but seem content being led by automation into anti-competitive behaviour or price gouging.
The Competition Commission of India’s (CCI) recent study on AI and how it could endanger fair competition makes interesting observations and suggestions, but the most remarkable of these is its call for market players to self-regulate.
Globally, the market for AI tools is showing winner-takes-all characteristics. Adoption entails concentration risks too. Aggressive AI deployment by large players, especially those with deep pockets and influence, could erect entry barriers, reduce competition and gain the dominance needed to overcharge.
A gold rush is underway for vast data-sets on customer preferences and behaviour for AI to study. As this takes heavy investment in infrastructure, technology and expertise, big players are far better placed to keep rivalry at bay.
“Removing these barriers is imperative to create a level playing field, encourage the entry of new players and stimulate competition in the AI marketplace," notes the CCI report.
‘Algorithmic collusion’ is another risk identified by its study. Even without human approval, AI tools across a market could read patterns and work out sneaky ways to cartelize and artificially inflate prices.
While the government is working on some solutions, such as its effort to grant universal access to critical data resources, the CCI report exhorts AI adopters to “self-audit" their AI systems to ensure they abide by India’s competition law. This has an air of dereliction about it.
Businesses, like all economic agents, respond to incentives and penalties. The CCI’s advice, alas, only amounts to moral suasion. It cannot induce anyone to self-regulate. True, AI poses a particular problem of opacity, as toolmakers are secretive about AI feeds and users do not know what makes it tick. But all our regulators need to adapt.
The rapid pace of AI uptake—outlays are expected to multiply five times over the next five years—requires them all to upgrade their capabilities. This new paradigm must not be left untackled.
On its part, the CCI can’t escape the need to issue guidelines—after seeking stakeholder inputs—that explicitly lay out AI guard-rails and penalties for their violation. Of course, such rules would need to be kept flexible and adaptive to evolving risks. But we need clarity right away, given how AI might already be swaying business decisions like retail pricing.
Unregulated AI deployment could be a slippery slope. Imagine ‘dynamic’ airfares getting personalized, with AI scraping our personal data to assess just how badly we need to fly—it’s a formula for extortionary ticket prices. The CCI should take the lead in defining how AI should be deployed so that it serves the common man.
topics
