Who is Manu Chopra? 27-year-old Indian on whom Google and Microsoft are relying for their AI success

Generative artificial intelligence like ChatGPT created a lot of buzz this year with companies rushing to capitalize on the growing demand and make their products better with accurate and reliable data. The frontrunners in the AI race like Microsoft and Google understand the value of quality data to make their AI story a success and 27-year-old Manu Chopra’s startup Karya has the answer to their data problems.

Headquarters in Bengaluru, Karya was founded in 2021 and successfully diversified the linguistic diversity of its workforce by focusing on recruiting individuals from rural regions who might not typically be engaged in such roles. Their application offers its services offline and offers vocal assistance to accommodate individuals with limited literacy skills.

Who is Manu Chopra?

Manu Chopra graduated in Computer Science from Stanford University in 2017, where he co-founded CS+Social Good, which was a student group focused on technology and impact. The 27-year-old is working on tackling extreme poverty in rural India by providing people with dignified digital work. Manu Chopra’s website claims that in the past 1.5 years, his work has moved over 1,00,000 rural Indians out of extreme poverty.

He also taught multiple courses focused on technology with a positive impact at the Computer Science Department at Stanford University. In these classes, students have undertaken projects that have had a positive influence on more than 30 million individuals across 15 different countries worldwide, his website said.

Through Karya, Manu Chopra is not just fulfilling the high demand for accurate data in the AI market, but also changing the economics for data annotation workers. Karya hires people from rural India, mostly women, and pays them over 20 times the minimum wage for their services which include gathering text, voice, and image data in India’s vernacular languages.

While speaking with news agency Bloomberg, the Stanford-educated computer engineer said that “every year, big tech companies spend billions of dollars collecting training data for their AI, so poor pay for such work is an industry failure.”

The tech giants are showing considerable interest in Karya’s for its data needs as Microsoft used the startup for the acquisition of regional speech data for its artificial intelligence products. As AI chatbots have several biases due to the feeding of large language models, the Bill & Melinda Gates Foundation is working with Karya to reduce gender-related biases in the data.

To aid Google’s expansion plans

Google is working on its expansion plans in India and aims to build an AI-generative model for India’s 125 languages. The company is relying on Karya and various other local collaborators to amass speech data across 85 Indian districts.

“India is the first non-Western country we are doing this in, and we are testing Bard in nine Indian languages,” Manish Gupta, head of Google Research in India told Bloomberg, referring to the company’s AI chatbot. “Over 70 Indian languages spoken by over a million people each had zero digital corpus. The problem is so stark.”

The report mentioned that in India, more than 32,000 crowdsourced laborers have utilized the application, completing a staggering 40 million compensated digital assignments, including tasks like image recognition, contour alignment, video annotation, and speech annotation.