18. Bonus Episode: Wadhwani AI Food Lab Talk podcast

1y ago 24:32

内容由Food Lab and Michiel Bakker提供。所有播客内容（包括剧集、图形和播客描述）均由 Food Lab and Michiel Bakker 或其播客平台合作伙伴直接上传和提供。如果您认为有人在未经您许可的情况下使用您的受版权保护的作品，您可以按照此处概述的流程进行操作https://zh.player.fm/legal。

Wadhwani Institute for Artificial Intelligence (Wadhwani AI) has been actively building and deploying AI solutions for underserved populations in developing countries since 2021. As a Google.org recipient, they are using AI to bring the world’s agricultural expertise to every farmer, protecting critical crops and mitigating the threat of hunger for billions of people. On this special episode of Food Lab Talk, Michiel speaks with Aditya Nayan and Soma Dhavala from Wadhwani AI to discuss the importance of understanding the audience and building trust when implementing novel programs, demonstrating value before scaling, and “open sourcing” innovation to open the door for other changemakers.

Wadhwani AI: “We're always inclined to think, okay, what is the specific problem statement that you're solving, how you're solving it and all that. But we're obligated to look at the larger system and see the value. So even in the case of CottonAce, there is as much value in that farmer getting the advisory as there is that the ministry can get a sense that this is the kind of infestation that's happening in these places and I can know that in real time. That's a starting point so that a year, two years down the line, I can create a predictive system which will help me prepare better. That's something that we didn't know in 2019 when we won the Google AI Impact Challenge. We didn't know this part of it. So it's this system thinking that where innovation also has to go and how things that we do maybe on the field, how they plug back in into the larger system is very important.” - Aditya Nayan

01:11 Intro to Wadhwani AI

03:36 How AI can add value in a rural environment

05:22 Why education is critical to adoption of novel technologies

06:19 How smartphones make AI and data accessible

07:17 Proof of concept with CottonAce

09:59 Scaling CottonAce principles to more crops

10:54 Building trust in technology through community engagement

12:35 Considering reliability and safety when developing new technologies

13:12 Open sourcing and “open syncing” for datasets, models, and innovation

15:20 Data is relative - how to determine impact

17:15 Measuring direct impact vs. indirect network effects

19:47 Why systems thinking is important for problem-solving

22:07 Takeaways for changemakers

Links

Keep in Touch

Subscribe, rate, review the show at foodlabtalk.com

Follow Food Lab talk on YouTube and LinkedIn

*The views expressed by the guests in this podcast don't necessarily represent the host’s views, nor those of his employer.

36集单集