Indian unicorn Ola has recently launched its AI cloud platform for developers, aiming to provide an alternative to OpenAI’s ChatGPT. The platform, named Krutrim after the Hindi word for “artificial,” was introduced in February with the Krutrim chatbot. However, the chatbot received mixed feedback from users due to erroneous and misleading responses. Ola’s founder Bhavish Aggarwal unveiled the AI cloud platform on May 2nd, offering models as a service and GPU instances for developers to fine-tune and deploy large language models. Despite Aggarwal’s emphasis on AI being the soul of the platform, the model’s quality has been criticized by users.
The platform features a comparison tool to evaluate model responses between Krutrim’s proprietary model, Meta’s LLaMA 3, and Mistral. Ola aims to provide a comprehensive AI platform for Indian developers, including custom silicon and a marketplace of AI apps for SaaS customers. During the launch event, Aggarwal criticized global hyperscalers like Google and Microsoft for not subsidizing their cloud platform costs for Indian developers, claiming Krutrim would be the most affordable platform in the market. However, Ola’s ambitious vision to deliver a complete AI platform remains questionable in the face of industry giants with dedicated research teams and proven track records.
While Ola’s ambitions are commendable, building a complete AI stack from scratch is a daunting task. Most companies in the AI landscape focus on specialized aspects of the technology stack, leaving infrastructure and platform development to cloud providers like Google, Microsoft, and AWS. Ola’s claim to provide affordable AI cloud services for Indian developers without a proven AI research record raises doubts about the platform’s competitiveness. Jaspreet Bindra, a leading voice in generative AI, highlighted the challenges of designing and delivering a comprehensive AI platform without prior experience in AI research.
A personal experience with Krutrim’s AI cloud platform revealed shortcomings in the model’s quality, lack of technical details, and inadequate documentation and APIs. The platform did not live up to the promises made by Ola’s founder and failed to impress Indian developers. While India needs AI solutions tailored for its diverse audience, Krutrim’s limitations hinder its potential to rival established players like OpenAI, Anthropic, and Mistral in the AI cloud space. Ola may need to refine its strategy and focus on key strengths to gain credibility among developers and fulfill its promise of disrupting the AI cloud industry in India. Until then, the platform’s journey to deliver on its ambitious goals remains a challenging one.
