Google and Accel India Invest in 5 Innovative AI Startups

A Rigorous Selection to Avoid Superficial Solutions

In the ever-evolving world of artificial intelligence startups, many ideas often emerge as mere "wrappers." These solutions simply overlay AI features on existing models without delivering true innovation. In light of this trend, investors are becoming increasingly cautious, seeking to avoid projects that could quickly become obsolete as AI model creators enhance their offerings.

When evaluating over 4,000 applications for their joint acceleration program, Google and Accel India found a predominance of these "wrapper" ideas. However, none of them were selected among the five startups chosen for the current cohort, as explained by Prayank Swaroop, partner at Accel, in an interview with TechCrunch.

An Ambitious Program to Support AI Innovation

Launched in November, the Atoms program from Google and Accel focuses on supporting early-stage startups developing AI products with a connection to India. The startups chosen for this cohort will benefit from funding of up to $2 million from Accel and Google's AI Futures Fund. Additionally, they will receive up to $350,000 in cloud credits and AI computing from Google.

"Wrappers" Dominate Rejected Applications

Approximately 70% of the rejected applications were "wrappers." These startups merely overlaid AI features, such as chatbots, on existing software without reinventing workflows through AI, Swaroop noted. Many other applications belonged to already saturated sectors, such as marketing automation and AI-based recruitment tools, where innovation is rare and differentiation is challenging.

A Growing AI Ecosystem

This phenomenon is not surprising, as the program received nearly four times more applications than previous Accel cohorts, with strong participation from first-time founders. The expanding AI ecosystem in India remains predominantly focused on enterprise applications. About 62% of submissions concerned productivity tools, and 13% were related to software development and programming, meaning that roughly three-quarters of the applications targeted enterprise software rather than consumer products. Swaroop hoped to see more ideas in the fields of health and education.

Startups Aligned with Google's Expectations

Jonathan Silber, co-founder and director of Google's AI Futures Fund, emphasized that the five selected startups closely align with areas where Google anticipates deeper adoption of AI in the real world. The program does not require startups to exclusively use Google's models, Silber clarified, noting that many companies combine multiple models according to their needs. The goal is to gather feedback on the performance of Google's models in real-world applications.

A Virtuous Cycle for AI Innovation

Insights gathered from these startups will be passed on to the Google DeepMind teams to improve future models, thereby creating a virtuous cycle between startup experimentation and AI development. "If a company is using an alternative model, it means Google has work to do to build the best model on the market," Silber told TechCrunch.

The Selected Startups

The five startups selected this year are:

• K-Dense, which is developing an AI "co-scientist" to accelerate research in fields such as life sciences and chemistry.
• Dodge.ai, which designs autonomous agents for enterprise ERP systems.
• Persistence Labs, which focuses on voice AI for call center operations.
• Zingroll, which is building a platform for AI-generated films and shows.
• Level Plane, which applies AI to industrial automation in automotive and aerospace manufacturing.