The Indian Space Research Organisation (ISRO) has indeed embarked on a preliminary study assessing the feasibility of establishing in-orbit Artificial Intelligence data centres, as confirmed by the Department of Space in response to a query in the Rajya Sabha.

This initiative reflects a strategic push towards edge computing in space, where satellites could process and store vast amounts of data directly in orbit rather than transmitting raw information back to Earth. Such an approach promises to alleviate bandwidth constraints and reduce latency for time-critical applications, particularly in satellite imaging and communications.

Preliminary evaluations by ISRO indicate that the concept is technically viable, leveraging advancements in onboard processors, satellite hardware, and solar power systems. The organisation is now conceptualising a proof-of-concept system capable of handling computation and storage in orbit. 

This builds on existing capabilities in onboard data handling, allowing satellites to reconfigure themselves dynamically for enhanced flexibility.

The drive stems from global data explosion challenges, where traditional Earth-based data centres grapple with power demands, cooling needs, and transmission delays. In space, uninterrupted solar energy and the vacuum's natural cooling provide distinct advantages, positioning orbital AI data centres as a transformative solution. ISRO's study aligns with this trend, aiming to process mission data closer to its source for efficiency.

However, formidable engineering hurdles persist. Space electronics must endure extreme radiation, severe power limitations, and thermal fluctuations in a vacuum environment. Security against cosmic threats and potential adversarial interference adds another layer of complexity, demanding robust radiation-hardened designs and secure orbital architectures.

Launch costs, though declining with reusable rocket technologies, remain a barrier to scaling such systems. High-bandwidth communication links between satellites and ground stations are essential yet challenging, requiring optical inter-satellite networks for terabit-per-second throughput. Orbital debris management and international regulatory approvals for large constellations further complicate deployment.

Globally, ISRO's efforts mirror initiatives by tech giants and agencies. Google’s Project Suncatcher envisions solar-powered AI satellite clusters in low-Earth orbit by 2027, using Tensor Processing Units for distributed computing. Companies like SpaceX, Blue Origin, and start-ups such as Aetherflux are prototyping similar "floating supercomputers," highlighting a competitive race India seeks to join.

For ISRO, this could revolutionise missions like earth observation, where petabytes of imagery overwhelm current downlinks. Onboard AI would enable real-time analysis, transmitting only insights—such as disaster alerts or crop health metrics—thus optimising resources. Communication satellites could benefit too, with in-orbit reconfiguration supporting dynamic bandwidth allocation.

Integration with India's broader space ambitions is evident. The Gaganyaan human spaceflight programme and growing private sector involvement via IN-SPACe could accelerate hardware maturation. Partnerships with domestic firms experienced in aerospace electronics, like those in Bengaluru's ecosystem, may prove pivotal.

Economic viability hinges on falling launch prices to around $200 per kilogram, a mid-2030s target with fully reusable systems. ISRO's cost-effective launchers, such as the SSLV and future medium-lift vehicles, position India favourably. Pilot projects might start with smallsat constellations demonstrating AI workloads.

Radiation mitigation strategies, including shielded processors and error-correcting algorithms, are critical. Solar arrays must deliver reliable power across eclipse phases, possibly augmented by advanced batteries. Thermal management in orbit demands passive radiative cooling, eschewing Earth's fans and water systems.

Data sovereignty gains prominence too. Orbital centres could process sensitive defence or strategic data indigenously, reducing reliance on foreign clouds. This aligns with India's Atmanirbhar Bharat vision in space technology, enhancing national security.

Long-term, scalability involves modular satellite swarms in sun-synchronous orbits for perpetual sunlight. Formation flying—keeping units kilometres apart—enables high-speed laser links, forming a resilient mesh network. Redundancy against failures ensures mission continuity.

ISRO's timeline remains exploratory, with conceptual designs underway post-feasibility confirmation. Union Minister Jitendra Singh noted AI integration across ISRO projects where feasible, subsumed in existing budgets. A prototype launch could materialise within 3-5 years, contingent on funding.

This endeavour underscores India's leap from launch services to space innovation leadership. By pioneering in-orbit AI infrastructure, ISRO not only addresses terrestrial limitations but also carves a niche in the global space economy, projected to exceed $1 trillion by 2040. Challenges abound, yet the potential for low-latency, green computing from space heralds a new era.

Agencies