From Prototype to Impact: Rethinking AI for Animal Welfare

At the Upadhyaya Foundation’s ‘AI for Animal Welfare: From Compassion to Action’ Demo Day, held on 7th March 2026 in Mumbai, developers, nonprofits, funders, and ecosystem stakeholders came together to explore how emerging technologies can be applied to animal welfare in practical, real-world ways. Organised in partnership with Electric Sheep and Open Paws, the event brought together a select cohort of AI teams working on solutions across wildlife protection, community animals, farmed animals, and rescue operations.

From AI-powered investigation tools and facility mapping to rescue coordination systems and behaviour change technologies, the solutions showcased were designed to integrate with NGO workflows, strengthen decision-making on the ground, and move from pilot to implementation with the right support. The room reflected this intent, bringing together developers, mentors, philanthropists, and NGOs positioned to adopt and scale these innovations.

To set the context for the day, Sam Tucker-Davis, Founder of Open Paws joined Ambika Chandra, Communications Manager at Upadhyaya Foundation in conversation. Their discussion explored how AI is already being used across the sector, the structural challenges that remain, and what it will take to move from early-stage prototypes to solutions that can scale and create meaningful impact.

Ambika: To kick things off, you’ve worked with hundreds of developers and activists to build AI tools. What are some real-life examples where AI is already helping animals?

Sam: I think corporate campaigns are a very interesting example here. This is especially relevant right now because 2025 was a year when many organizations had made commitments. For example, large supermarket chains committed to going cage-free, and 2025 was the year those commitments were meant to come into effect.

But in many cases, that didn’t happen.

So there’s a huge amount of potential to use AI as a tool for scaling outreach. This is typically a very time-consuming process that requires a lot of volunteer energy to get companies to follow through. AI can help by identifying pressure points in campaigns and supporting outreach by generating initial drafts for social media, emails, and other communications.

More broadly, beyond corporate campaigns, almost everything the movement does involves some form of outreach. Whether it’s engaging government representatives, company leaders, or the public, outreach is central.

Historically, we’ve had a trade-off. We could either send generic messages at scale or create highly personalized outreach for a few individuals. We couldn’t do both.

AI changes that. It enables hyper-personalized outreach at scale, which is a very powerful use case for the movement.

Ambika: Absolutely. And as someone working in marketing and communications, that’s such a strong use case for AI.

I also love that it’s not just organizations or NGOs that can use these tools, but also individuals who care about what they consume or want to take action. That’s really powerful.

We’ve also spoken about how innovation for animals won’t come only from technologists. Coders are doing incredible work and are integral to the process, but what role do you see NGOs, funders, and other stakeholders playing in using AI for animals?

Sam: That’s a great question.

One of the most helpful things for developers has been understanding the real challenges organizations face in their day-to-day workflows.

The best ideas don’t come from an NGO saying, “We want this specific tool.” They come from conversations about underlying problems.

When you understand those problems, you can identify solutions that organizations may not have even realized were possible with AI.

At Open Paws, whenever we work with a new organization, the first step is an audit. We ask staff about their workflows, what tasks are repetitive, and what takes up time without contributing to core work.

That’s really the key. There needs to be a bridge between the technical side and real-world use cases.

Ambika: Absolutely. It just shows that meaningful change requires many stakeholders to come together and play a role.

For organizations here who are curious and want to start using AI, what should that first step be?

And I’m asking this as someone who isn’t necessarily very tech-savvy. If I wanted to implement AI in my work, where would I begin?

Sam: The first step is to reach out. We offer free AI audits, and since we’re a nonprofit ourselves, that’s not a paid service.

Another option, if you’re hesitant to reach out, is to use AI itself. You can ask tools like

ChatGPT, based on what you know about your work, how you might start using AI.

It might sound a bit meta, but you’ll get useful starting points.

Ambika: I love that. And yes, sometimes it’s hard to know who to reach out to, but we all have access to these tools, so why not start there.

Coming to some of the challenges in building tech solutions, high-quality data is often a constraint in the animal welfare ecosystem, especially in certain focus areas.

How do we overcome this challenge?

Sam: That’s a great question. I don’t think the issue is a lack of data. The issue is that data is siloed and fragmented.

For example, when an organization runs a campaign on social media, a lot of high-quality data is generated, like engagement metrics. But there’s no system for sharing that data across organizations to generate movement-level insights.

At Open Paws, one of the first things we did was create data-sharing agreements with about 50 animal advocacy organizations. That allowed us to build shared datasets and train our models more effectively.

At a higher level, the answer is collaboration. And that’s true for many of the challenges we face.

Ambika: Absolutely. Collaboration helps not just in building AI tools, but in solving many broader challenges as well.

We’re here today for this demo day, and we’re about to see some exciting prototypes. But how do we move from early-stage ideas to solutions that are scalable, replicable, and impactful?

Sam: The process has to be iterative and involve end users from the beginning.

You don’t build a product and then test it later. You involve users at every stage.

That’s why we start with problem statements from advocacy organizations. We identify real needs, and as we build, we continuously involve NGOs to refine the solution.

Many organizations don’t have strong technical capacity, so tools need to be extremely user-friendly. The only way to achieve that is to test them with real users and iterate.

Ambika: That makes a lot of sense.

For the audience, as we listen to the solutions today, what should we look for to understand whether something can scale and sustain?

Sam: Great question.

First, some context. The past four weeks have been a pre-accelerator, and the winning teams will go into a full accelerator with three more months of support. So these solutions are still early-stage.

In general, for scalability, look for products where the workflow remains consistent, but the inputs and outputs can vary widely.

That’s something AI enables. Earlier, solutions had to be built for very narrow use cases. Now, the question is how many related problems a solution can address.

A good analogy is large language models. They follow a consistent process, but you can give them very different inputs and still get useful outputs.

Ambika: That’s a helpful way to think about it.

And just to wrap up, looking ahead, what potential do you see for AI to transform the animal welfare ecosystem?

Sam: It’s difficult to predict exact timelines, but we can look at trends.

Earlier, AI functioned like a junior assistant, handling small, well-defined tasks with a lot of instruction.

Now it’s evolving into something closer to a junior employee. And eventually, it could function like a manager overseeing a team.

This means we need to rethink how we operate as a movement. We’ve historically worked with limited resources, trying to address large problems with very little capacity.

As technology improves, many of those constraints will reduce.

So individual activists need to start thinking like managers. Managers need to think like organizational leaders. And organizations need to think at the level of the entire movement.

We need to keep expanding our sense of what is possible.

Ambika: Thank you so much, Sam. That was really insightful.

It’s given us a lot of context on what’s possible, why we’re here, and how we can use the ideas from today to think more broadly and be more ambitious in our work.