Indian AI startups tackle real-world problems amid funding and computing challenges

6 Min Read
indian-ai-startups-real-world-problems-commercialisation-challenges

Indian AI startups are increasingly applying artificial intelligence to solve complex problems across property, healthcare and infrastructure.

Rather than competing directly in frontier AI model development, these companies are focusing on practical applications. They are building tools around fragmented government records, multilingual data and public services.

One example is Landeed, founded by Sanjay Mandava after a personal experience with India’s property documentation system.

In 2020, Mandava travelled from Chicago to Chennai to resolve a family land dispute. He discovered that establishing ownership required several documents beyond the original sale deed.

The experience led him to develop Landeed, an AI-powered platform for accessing land records and checking property titles. The platform operates across 24 Indian states.

Landeed uses AI to simplify property verification

India’s property ownership system places significant responsibility on buyers to establish clear title. Unlike systems where governments guarantee ownership, a registered sale deed alone may not establish an undisputed title.

Buyers may also need property tax receipts, building certificates and land survey records. Consequently, verifying ownership can involve multiple government departments and databases.

Landeed brings together information from registration offices, land revenue departments, municipal databases and survey records. It uses geospatial technology to connect registered properties with official identification numbers and ownership details.

Its AI layer extracts information from documents and identifies inconsistencies. This can reduce checks that traditionally take lawyers and property consultants several days.

The company says its repository contains around 1.8 billion title documents. Millions of individual and enterprise users now access the platform for property-related transactions.

Indian AI startups target infrastructure and healthcare

Beyond property technology, Indian AI startups are developing applications for civic infrastructure and public healthcare.

For example, Bengaluru-based entrepreneur Gaurav Sen has developed an AI application to identify road damage. The system uses dashcam footage, satellite navigation and computer vision to detect potholes.

It can also identify the civic authorities responsible for addressing reported road damage. Sen told the BBC that municipal corporations could deploy the technology across public buses and taxis.

This would provide updated information about road conditions across urban areas. Meanwhile, several other platforms are also using technology to map and report damaged roads.

In healthcare, Fractal Analytics has deployed its AI application, Vaidya, in hospitals operated by Mumbai’s civic authority.

The platform provides simplified summaries of medical reports in multiple languages. Additionally, it helps users identify nearby healthcare facilities based on their requirements.

Doctors can access AI-assisted patient information before consultations. The system also provides hospital administrators with aggregated, anonymised data to monitor public health trends.

These applications demonstrate how companies are adapting AI to local requirements. They are addressing challenges involving language diversity, fragmented information and limited public-service capacity.

AI startup funding rises despite commercialisation challenges

The expansion of these applications has attracted investor interest. Indian AI startups raised more than $1 billion during the first nine months of the year.

Their funding increased 265 per cent year-on-year, according to startup information platform Inc42. This growth came despite relatively subdued fundraising across the wider startup ecosystem.

However, converting pilot projects into sustainable businesses remains difficult.

Fractal Analytics co-founder Srikanth Velamakanni told the BBC that the cost of developing and operating AI products can exceed what Indian customers are willing to pay.

The computing requirements also remain comparable across markets. However, customers in the United States generally have greater purchasing capacity than Indian buyers.

This creates a commercial challenge for companies developing expensive AI solutions for price-sensitive markets.

GPU shortages and regulation constrain AI development

Access to computing infrastructure is another major concern for Indian AI founders.

Mandava said Landeed faced difficulties securing graphics processing units (GPUs), which support the training and operation of AI models.

The government has announced plans to provide 38,000 GPUs through its IndiaAI Mission. Nevertheless, demand for computing capacity continues to exceed available supply.

Furthermore, regulatory requirements and administrative delays can complicate business expansion.

Companies also face challenges in hiring experienced talent and converting successful government pilots into recurring contracts.

These constraints could limit the long-term viability of early-stage AI products. Experts have also cautioned that basic applications may become easier to replicate as technology advances.

Sen acknowledged that his pothole-detection platform would need greater sophistication to remain competitive over time.

Indian AI ecosystem seeks support to compete globally

India’s AI development is increasingly centred on applications that address domestic economic and social needs. However, scaling these solutions will require reliable computing infrastructure and commercially viable business models.

Founders are seeking access to technology and resources rather than protection from global competition.

Mandava has called for stronger support to help Indian companies compete with international players. His concerns reflect the wider challenge facing the ecosystem.

The next phase of growth will depend on whether startups can move beyond promising pilots. They will need to build scalable products, secure paying customers and reduce dependence on expensive external technology.

TAGGED:
Share This Article
Leave a Comment

Leave a Reply

Your email address will not be published. Required fields are marked *