In Conversation: Sapien Head of Sales Chris May Discusses AV Models, Data Demands, and More

We've seen massive changes from the AI industry since Sapien’s founding, with customers demanding more specialized expertise, and our decentralized global labelers now able to take full advantage of our platform. To understand how Sapien is adapting to these changes, we spoke with our Head of Sales, Chris May, about the latest trends driving demand in the industry. From the autonomous vehicle sector to the rise of large language models (LLMs) requiring domain-specific fine-tuning, we explored how Sapien is positioning itself to meet these shifting needs.

‍

1. Which industries are currently driving the highest demand for data labeling services, and how do you see that evolving over the next year?

The demand for data labeling is definitely being driven by the autonomous vehicle industry. They’ve always needed a centralized, high-quality labeling workforce to train their systems, and that’s not going away. But what’s really interesting is the shift we're seeing with LLMs. More and more companies are developing large language models, and they’re not just looking for generic data anymore—they want domain experts who can fine-tune these models for specific use cases.

So, while autonomous vehicles will continue to dominate in terms of volume, I think we’re going to see a growing demand for specialized labeling. And the only way to meet that need is through a verified, decentralized network of experts who can bring that level of nuance. It’s definitely an exciting evolution.

‍

2. What specific challenges are you discovering or finding when serving high-demand sectors, and how are those industries’ needs influencing your sales strategies?

One of the big shifts we're noticing is that a lot of the general-use AI models—think the ones trained on broad, massive datasets over the last decade—are hitting a maturity point. They're great for covering a wide range of needs, but enterprises are now leaning towards something more precise. We're seeing a pivot to industry-specific models, whether it's in legal, healthcare, finance, or other specialized fields.

What that means for us, from a sales perspective, is that we can’t just talk about AI in broad strokes anymore. We have to dive deep into these niches and understand the particular problems they’re facing. Our go-to-market strategy is increasingly about building tailored collateral and developing tools that speak directly to those unique challenges. It’s not just about selling a solution; it’s about showing that we truly get their world and can make a difference.

‍

3. Where do you see the most significant growth opportunities?

Highly skilled workforce contributing to model training to create models with domain expertise. We’re at a point where tools like ChatGPT can produce outputs that extract data from a public domain that is fairly accurate and useful. (ex: give me a step-by-step guide on how to make a beef wellington); but what if LLMs can start giving extremely accurate answers to even more technical answers on legal, healthcare, and financial inquiries where you would only be able to find data from speaking with a highly trained attorney, physician and financial expert?

‍

4. What are some of the key trends in the industry, particularly in sectors like automotive and data collection? How are you adapting to meet those demands?

Data collection has been, in terms of the number of projects, the largest requested service at Sapien. Enterprises are building AI models that can understand sophisticated inputs such as multiple languages in different accents, business conversations, music, and other forms of data.

In order to meet those demands, Sapien has amassed a global network of experts that can be ready to take skills assessment tests to be verified and start imparting their domain knowledge into training models.

‍

5. Which regions or markets are showing the most growth in demand for data labeling, and what industries in those areas are driving that demand?

We were surprised at how big the labeling and data collection demand is in Asia, specifically China. Sapien has partnered with companies like Alibaba, TAL, and many others to assemble global expertise to help with projects such as TTS, ASR, and LLM fine-tuning.

In terms of industry, Autonomous driving is still the largest vertical, but we have been getting significant request for projects such as Wake-work collection, Image and audio clip collection.

‍

6. In sectors where accuracy is extra critical, how is Sapien balancing the need for speed in data labeling with the demand for high-quality, precise annotations?

One of the differentiators with Sapien’s solution is that we custom design labeling software for our customers. A large part of the customization comes from adding automation in the tagging module and designing the QA workflow.

We can integrate pre-built models for faster tagging and design custom QA processes to ensure data is being sampled and checked according to customer requirements.

‍

7. What role do automation and AI play in meeting the current demand, and where do you still see a strong need for human expertise in the data labeling process? How does Sapien balance this now and in the future?

Models are just going to get better and more accurate, but not without domain expertise and vast amounts of input training. In order for this to happen, a solution is needed to meet both demands from the supply and demand side.

At Sapien, we are listening to enterprises all around the world, and hearing that there is a tremendous amount of demand and willingness to pay for domain expertise; and from the supply side, we have people all around the world looking to impart language, legal, healthcare…etc expertise that meet these demands.

Sapien is providing a platform to combine the two worlds with both Web2 and Web3 solutions, which will allow us to potentially create the world’s largest AI ecosystem.

As demand for high-quality, precise data labeling increases, Sapien remains focused on delivering customized data labeling and data collection services that combine human expertise with automation. With our global network of labelers and gamified platform, we're leading the way in creating an ecosystem that bridges supply and demand for data, to stay ahead in a rapidly changing industry.