Introduction
VC investments in AI have seen huge surges in the past few years. Early-stage AI startups cumulatively raised close to $560 million in funding. Startups such as Krutrim, Sarvam.AI, and Mad Street Den have raised significant funding from VC investors. Krutrim was valued at $1 billion in its last funding round. This showcases the potential of AI startups to provide huge exits to investors.
This blog will explore the key drivers, trends, and challenges faced by AI startups in India.
Why are VC investors betting big in India?
VC investors are betting big on Indian AI startups, mainly because the usage of AI in industries is expanding. These areas include content, gaming, astrology, health, education, and law.
VC investors are extremely interested in India’s AI industry due to the realization among founders and experts that general foundational models cannot constitute a successful business model for a startup. Furthermore, developing a model is expensive and requires a lot of investment.
In India, this is why there is a rise of Vertical AI startups that build software that solves industry-specific challenges, offering additional capabilities over and above the foundational models. SpotDraft (Legal Field) and Track3D (Construction) are examples of Vertical AI startups. Vertical AI startups are gaining traction even in competitive markets due to their ability to laser-focus on solving specific problems.
While the growing demand for AI solutions is a major factor in driving VC interest, government policies towards AI startups further enhance their attractiveness. The Indian Government increased the budget of the AI Mission by allocating Rs 10,372 crores for building computing infrastructure through public-private partnerships. India’s new Draft Deeptech Policy 2023 aims to accelerate technological growth and global competitiveness.
These policies not only offer startup funding but offer opportunities for research, access to unified data sets, and promotion of AI applications for startups tackling socio-economic challenges. These benefits make the launching of AI startups easier and offer easier access to resources for building an AI startup. Investor confidence is boosted.
In addition to all of this, India’s burgeoning talent pool is one of the reasons for VC investors. Companies and the Government are launching initiatives for continuous upskilling, and it is crucial to ensure this. Otherwise, there would be a disequilibrium in the demand-supply of talent, which is crucial to ensuring innovative startups.
Challenges faced in scaling AI startups
1) Data Privacy Regulations
AI startups in India have to comply with the regulations set forth in the Digital Personal Data Protection (DPDP), 2023. This is a huge challenge for such startups because access to high-quality data is vital for training models. Using publicly available data could conflict with existing copyright and the DPDP bill. Inaccuracies and bias in the GenAI applications involving hiring, marketing, and insurance claims pose critical risks. Questions relating to whether the data fiduciary or the startups developing the model are crucial. AI Regulations are still in the nascent stage and, therefore, have to be navigated by startups through legal counsel specializing in AI.
2) Limited Talent Pool
AI being a highly specialized and niche field, the number of professionals is limited. Sourcing the right people with the correct skills is crucial for building an AI startup. The technical expertise required for building a model is high. Initiatives like the government’s YUVAi (Youth for Unnati and Vikas with AI program) and companies like Tata Consultancy Services, Wipro, and HCLTech continuously upskilling employees would help increase the talent pool.
3) Bias in Training sets
The data sets used for training AI models must be inclusive, as India’s diverse population requires attention to prevent discriminatory output in AI systems. Discriminatory output by AI systems could exacerbate the existing prejudices and biases in society. This problem is mainly faced by smaller startups, which often lack the resources to promote fairness and responsibility. This can be solved by using diverse datasets, including a wide range of scenarios and demographic groups. The models should be tested on a regular basis, and “Bias tests” can be conducted. These tests can be utilized to update the models and reflect the latest changes in society.
Conclusion:
AI startups in India are rapidly evolving due to a combination of factors mentioned above. They are poised to revolutionize India’s startup ecosystem, and the growth is driven by favourable government policies, increasing talent pool, and a shift towards Vertical AI startups.
While the potential for investment and innovation is immense AI startups must navigate challenges such as data privacy, stagnant talent pool and bias in training sets to ensure their success. Despite this, India’s AI startups continue to grow through upskilling initiatives and inclusive data sets, creating a ripe environment for innovation.
The future of AI startups in India is bright but the success of these startups will depend on them innovating responsibly and navigating these challenges successfully.
Author Bio
