Last week, President Joe Biden’s Executive Order set a new course for the United States (US) on Artificial Intelligence (AI) safety and security, signalling a turn towards more robust standards. The order commits the US to collaborate internationally on AI governance. Hot on the heels of this move, India joined 27 other countries in adopting the Bletchley Declaration, a commitment from the recent UK-led AI Safety Summit convened by Prime Minister Rishi Sunak, that acknowledges the need for a global
Last week, President Joe Biden’s Executive Order set a new course for the United States (US) on Artificial Intelligence (AI) safety and security, signalling a turn towards more robust standards. The order commits the US to collaborate internationally on AI governance. Hot on the heels of this move, India joined 27 other countries in adopting the Bletchley Declaration, a commitment from the recent UK-led AI Safety Summit convened by Prime Minister Rishi Sunak, that acknowledges the need for a global alliance to combat AI-related risks such as disinformation.
The exact shape of the proposed international cooperation on AI remains undefined in these proposals. This gap in clarity precedes India’s role in chairing the upcoming Global Partnership on AI (GPAI) Summit this December. The G7 nations and India, among others, formed the GPAI in 2020, which now consists of 29 jurisdictions. GPAI invites a spectrum of stakeholders, including academia, industry, and civil society, to the AI governance table. This sets the stage for a crucial question: In what ways can India shape the global governance of AI — a technology already interwoven into our daily lives?
To find the answer, we can look at the evolution of the governance of the internet itself. Broadly, the internet is divided between a services layer and an infrastructure layer. The former consists of applications and content and is regulated by States while the latter requires a more global management because it is interconnected. The conflict between a not-quite-perfect multi-stakeholder approach and sporadic State-controlled multilateralism has marked the history of standard-setting that governs the infrastructure layer.
A multi-stakeholder model has more or less dominated the governance of internet infrastructure since the 1990s. Organisations such as the Internet Corporation for Assigned Names and Numbers (ICANN), the Internet Engineering Task Force (IETF), and the World Wide Web Consortium (W3C) — none of them government bodies — are in charge of the internet’s technical backbone. Specifically, the Domain Name System (DNS) is administered by ICANN, the Transmission Control Protocol and the Internet Protocol maintained by IETF, and the Hypertext Markup Language developed by W3C.
This model’s limitation on direct State intervention has sparked repeated global contestation. For instance, ICANN is a private entity that operates under California’s jurisdiction. And the DNS regime that ICANN administers is like a centralised phonebook for the internet, which allows web browsers to connect with websites. Therefore, a key part of how the internet functions is managed by a business domiciled in America, which makes the Sino-Russian block particularly uncomfortable. Therefore, China has evolved an exceptional internet structure, which is a closed system that resembles an intranet, with Russia likely following suit. This invasive ringfenced model provides a glimpse into a dystopian alternative to a global internet.
In fact, China, Russia, and even West Asian countries have actively campaigned to transition internet governance from multi-stakeholder to multilateral organisations since the early 2000s. Central to their strategy is to bolster the role of the United Nations in managing the internet, to create a counterbalance to the domination of technical bodies they distrust. While the UN did create a Working Group on Internet Governance in 2004, this did not upend the institutional status quo. Instead, the world got an “Internet Governance Forum” (IGF) in 2005, which serves as a platform for dialogue, rather than regulation of the internet.
India has evolved from a proponent of multilateralism to a supporter of multi-stakeholderism within IGF discussions. The country is also fortunate to have a wealth of non-governmental expertise to contribute to digital policy, a resource that can greatly enhance our capability in international forums if nurtured well. This pool of expertise has actively contributed to significant shifts in domestic policies. It encompasses communities that have championed major initiatives like net neutrality and the right to informational privacy, leading to legal reforms. India can advance its values and interests via a truly inclusive multi-stakeholderism much further than it can through multilateralism.
Moreover, the trajectory of AI is even more unpredictable than that of the early internet and it is evident that such a powerful technology requires societal oversight. Neither governments nor technical communities should unilaterally shape our AI-mediated future. The societal impact of AI will ripple through as developers release new technologies from laboratories to the real world. These waves will wash over all prevailing assumptions about the future of work, the climate crisis, human development, and economic growth. Developers of these technologies simply cannot fully predict how societies will use, modify, and react to different applications. Hence, the participation of multiple stakeholders in AI governance is not merely beneficial, it is essential.
As India prepares to lead GPAI, it has a unique opportunity to embed a truly democratic multi-stakeholder ethos in AI governance from the start. Unlike the internet’s early days, there are no entrenched technical bodies to contend with and all GPAI members are like-minded. With close to a billion connected citizens, India represents a fifth of global internet users, who already experience AI in its nascent forms thanks to a burgeoning digital economy. Our country isn’t just another participant in a new global forum; it is poised to define AI governance in the days to come.
Vivan Sharan is a partner at Koan Advisory Group. The views expressed are personal
One Subscription.
Get 360° coverage—from daily headlines
to 100 year archives.
Archives
HT App & Website