Synthetic Respondents in Market Research: Risk or Reward?

Market research is all about understanding real people—how they think, feel, and behave. But what if those "people" weren’t people at all?

Enter synthetic respondents: AI-generated participants designed to simulate human survey responses. Sounds futuristic, right? Some hail this as the next big thing in research, offering speed, scalability, and cost savings. Others worry about authenticity, ethics, and whether AI can ever truly replace human insight.

So, what’s the reality? Are synthetic respondents an exciting innovation or a potential risk to data integrity? Let’s dive in.

What Are Synthetic Respondents?

Put simply, synthetic respondents are AI-generated personas that take part in surveys, focus groups, and other research methods—without a real person behind them.

These virtual participants are created using machine learning and behavioural modelling to mimic human responses, often based on vast datasets of real-world consumer behaviour.

The idea isn’t entirely new. AI-driven analytics and predictive modelling have been used in research for years. But instead of just analysing existing data, synthetic respondents actively participate, providing answers just like a human respondent would.

But can AI really replicate the complexity of human thought?

The Pros of Synthetic Respondents

There’s no denying that AI-powered respondents bring some big advantages to the table:

Cost Efficiency – No need to recruit, incentivise, or manage real participants. AI can churn out responses at a fraction of the cost.

Scalability – Need 10,000 survey completions overnight? No problem. AI respondents don’t sleep.

Flexibility – Synthetic samples can be tailored to any demographic, ensuring balance across age, income, and behaviour segments.

Speed – AI can analyse and respond instantly, slashing research turnaround times.

Bias Reduction – In theory, AI removes human biases—no social desirability effects, no memory errors. But is that really the case?

The Risks and Challenges

For all their potential, synthetic respondents come with big questions—and even bigger risks.

Lack of Authenticity – AI models rely on historical data. But can they generate new insights, or are they just repackaging old ones? Market research is about exploring what’s changing, not just what’s already known.

Oversimplification – Human decisions are complex, emotional, and often irrational. AI, no matter how advanced, follows patterns and probabilities—it doesn’t feel.

Ethical Concerns – Who’s ensuring that synthetic respondents are being used responsibly? Without transparency, clients may not even realise that their data is AI-generated.

Regulatory and Industry Standards – The MRS Delphi Report on AI warns that AI in research should align with ethical standards, particularly in how data is collected and validated​. The IQCS and MRS are actively working on AI guidelines, but the industry still lacks firm regulation.

Trust & Data Integrity – Will businesses trust insights from AI-generated respondents? If synthetic samples become widespread, could we see data dilution, where real human behaviour is no longer properly represented?

Debrah Harding, Managing Director of MRS, put it well:

"In order to create synthetic data, you still need really, really good human data."

AI can complement real research—but it’s not a replacement for genuine, human-driven insight.

Will AI Replace Human Respondents?

Probably not. At least, not entirely.

While synthetic respondents might work well for predictive modelling and early-stage concept testing, they struggle with:

Emotional and subconscious decision-making – Humans buy products based on gut feelings, nostalgia, and social influence—things AI can’t fully replicate.

Qualitative research – A virtual respondent can’t engage in a genuine conversation, pick up on tone of voice, or uncover those all-important ‘aha’ moments.

Cultural and social contexts – Trends, slang, and cultural attitudes shift fast. AI trained on past data might not keep up.

Harding sums it up well again:

“I don’t see synthetic data replacing these techniques. The mix will be different, but the need for real people will remain.”

At Angelfish Fieldwork, we’re committed to real respondents—people with lived experiences, emotions, and unique perspectives. AI is exciting, but when it comes to human insight, there’s no substitute for the real thing.

The Future: A Mix of AI and Human Insight?

Rather than choosing AI vs Humans, the future of market research will likely involve both:

✅ AI for large-scale data processing, predictive insights, and synthetic modelling
✅ Humans for qualitative depth, emotion, and real-world unpredictability

Brands will need to decide when AI makes sense—and when human respondents are irreplaceable.

Our take? When you want insights that truly connect with real people, you need real respondents. No AI can replace human stories, emotions, and experiences.

So, while synthetic respondents might be an interesting tool, for high-quality, trustworthy research, humans still have the final say.

What’s Your Take?

Would you trust AI-generated research results? Do you think synthetic respondents will help or hinder market research?

Interested in conducting research with real human respondents? Get in touch with Angelfish Fieldwork today!

Sources & Further Reading