By Gene Saykin, VP, Head of Quality Strategy at Toluna

AI is reshaping the market research industry by changing the way we collect and interpret data. It offers unprecedented speed, accuracy, and depth of insight, bringing incredible opportunities to businesses.

But just as a high-performance car requires quality fuel, AI needs high-quality data to produce reliable insights. Poor data quality (from incomplete, outdated, or biased data) can lead to flawed outcomes and misguided business decisions. Recognizing the importance of data quality is the first step to unlocking AI’s true potential. Equally important is using that power ethically and responsibly.

The dark side of AI

The fraudulent use of AI poses significant risks to data integrity. Malicious actors can exploit AI to manipulate data, creating fake survey responses, skewing sentiment analysis, spamming your company’s emails, scraping your entire company’s directory via LinkedIn, training AI LLMs to pretend to be specific B2B participants based on their LinkedIn profiles, reverse-engineering the algorithms you’re using for some of your quality checks, or generating synthetic data to alter research outcomes. These examples of the fraudulent use of AI are but a few; the list is much longer. This threat is becoming increasingly prevalent as AI tools become more accessible and sophisticated.

In addition, biased data threatens data integrity by embedding skewed perspectives, which result in flawed insights. Bias can be inadvertently introduced by researchers themselves, for instance, by selectively cleaning certain responses from datasets. When AI models are trained on biased data, the insights they generate often mirror these biases, skewing results in favour of overrepresented demographics. This can lead to misguided strategies that fail to resonate with the broader market, potentially alienating key customer segments.

Compounding this issue is the rise of AI-generated survey bots, which mimic human responses with increasing sophistication. These bots can closely replicate genuine responses, making it very difficult to detect fraud using traditional methods. They may vary their answers, adjust completion times, and even simulate realistic thought patterns, slipping past standard anti-bot checks. This complexity challenges researchers, who must now adopt advanced fraud detection techniques to filter out these responses and protect data integrity in large datasets.

The need for ethical AI practices

As AI is now playing a larger role in data analysis, ethical practices become critical to ensuring the collection of reliable, high-quality data and, as a result, trustworthy and actionable insights. This involves being transparent in how AI models are developed and used, as well as conducting regular audits to check for biases and other ethical concerns.

Ethical AI practices also include ensuring that AI systems are designed to avoid reinforcing existing biases or introducing new ones into the decision-making process. Regular audits of AI systems can help identify any unintended biases that may have crept in during the training phase due to unrepresentative or skewed datasets. By addressing these issues proactively, businesses can maintain the integrity of their AI systems and build trust among stakeholders who rely on AI-driven research to make informed decisions.

Ethical AI in action

The good news is that organizations are adopting several concrete measures to ensure ethical AI use in market research. These include training AI models on diverse, representative datasets and conducting regular audits to detect and address biases. Advanced fraud detection tools, such as behavioural analysis and real-time monitoring, are increasingly being used to identify and filter AI-generated survey bots and fraudulent responses. Synthetic data is also being ethically applied to enhance fraud detection without compromising sensitive information, while human oversight ensures AI-driven insights remain contextually relevant and trustworthy.

Transparency is another key ethical practice, with many companies documenting AI processes and sharing plain-language summaries with stakeholders. Industry collaboration and compliance with regulations helps uphold respondent privacy and informed consent.

At Toluna, we align our AI and data practices with ESOMAR and IA standards. These guidelines shape how we approach consent, data protection, respondent privacy, and methodological transparency.

Moreover, our AI systems are routinely audited to detect and correct bias, especially when training on large-scale datasets. We leverage diverse and representative training data and employ fairness metrics to evaluate model performance across different demographics. We ensure human oversight at every step: in model development, interpretation of insights, and resolution of ambiguous data quality flags.

To ensure data integrity, we implement robust fraud detection that combines AI-driven behaviour analysis, pattern recognition, and real-time validation. This helps us identify survey bots, duplicated identities, and synthetic responses before they influence insights. AI is at the heart of Toluna’s solutions, platform, and methodologies. We’re committed to ethically unlocking AI’s full potential through responsible innovation, robust governance, and a human-centric approach to data. This commitment ensures our use of AI drives results as well as upholds the standards our clients and industry trust us to maintain.

About Toluna

Toluna is the leading global research and insights provider that empowers clients to make smarter data-driven decisions. For 25 years, we’ve partnered with our clients to deliver greater business impact through our advanced platform, end-to-end solution portfolio, deep industry expertise, and expansive global first-party panel.

From DIY to full-service consultancy, our unique approach can be tailored to fit any time, budget, and resourcing requirements. From creative testing to brand tracking, and from packaging optimization to e-commerce understanding, Toluna’s range of solution suites provide brands with the insights required to drive growth, relevance, and profitability.

With over 40 offices across the globe, we deliver research in 70+ countries to the world’s leading brands. Recognized as a top innovative company by GRIT in 2024, we are at the forefront of insights innovation and excellence.

This article was first published in the Q2 2025 edition of Asia Research Media