The pace of change in the market research and insights industry is staggering. With AI-powered tools, synthetic data, and automated research platforms entering the scene, the boundaries between traditional research and adjacent industries are blurring. What people want and how we understand it is evolving fast. So, what’s really going on in this era of disruption and diffusion?

Amara’s Law and the long-term impact of emerging technologies

Let’s start with Amara’s Law: We tend to overestimate the effect of a technology in the short term and underestimate it in the long term.

That pretty much sums up the state of the research industry. AI isn’t just knocking on the door of market research. It’s moved in, rearranged the furniture, sitting on the couch and watching the TV. It wants us to get in the kitchen and get cracking on dinner!

Back to reality, there’s a three-way tussle over insights spend between (i) Corporate IT, (ii) Research Technology (Res Tech) Platforms, and (iii) Traditional Research Consultancies (aka ‘Humans’, which is not to say corporate IT aren’t human too). Each camp is bringing its own ethos, tools and capabilities. These include AI-driven platforms, synthetic data, CDPs and increasingly sophisticated decision automation.

The machines are learning (and so are we).

The 2025 GRIT Insights Practice Report paints a clear picture:

• Internal research is on the rise

• AI is already embedded in both client and supplier-side workflows

• Smaller agencies are under pressure to adapt or perish in a tech-enabled ecosystem

And here’s the hard truth: human labor in research is being displaced.

Take RAG (retrieval augmented generation) as an example. By feeding AI a company’s best practices, research frameworks, and language templates, we’re enabling tools that can design studies, generate content, and answer business questions at scale, and with consistency. Or using Agential AI, with multi-agent processes, to automate and coordinate work step production across the research value chain.

Qualitative research at scale: AI moderators and automation

What about qualitative research, traditionally the “human” stronghold?

Say hello to AI moderators that can probe, nudge, and engage with thousands of participants at once. These tools are enabling scaled human engagement while dramatically lowering the cost of qual. The old barriers, like scheduling, transcription, and analysis, crumble in the face of intelligent automation.

Massive volumes of unstructured free text are now fair game. AI doesn’t just summarize; it discovers hidden patterns, correlations, and relationships that humans would miss.

Is primary research a sunset industry? Not entirely. But it’s no longer the only show in town.

We’re witnessing a shift from primary data collection to data enrichment. Platforms now blend survey data with behavioral insights, CRM records, third-party feeds, and even synthetic data - like Columbia Business School’s panel of digital twins. These 2,500 AI personas mirror real humans, trained on cognitive, attitudinal, and behavioral ground truth. In tests, these twins were able to replicate human survey responses with 85% accuracy. Scary? Maybe. Impressive? Absolutely.

Six hard truths about AI's impact on market researchers

Let’s call it what it is:

1. AI often eliminates the need for new research
2. Briefs are becoming less structured, more fluid
3. Speed and cost have redefined expectations
4. Brands are building internal AI tools - and using them
5. Traditional qual and quant are too slow
6. Human-led insight is no longer the default premium

The moat is gone. Models can reason, not just predict.

Multimodal Communication Protocol (MCP): The nervous system of AI agents

One acronym you’ll be hearing a lot more of: MCP: Multimodal Communication Protocol. Think of it as the nervous system that allows autonomous AI agents to communicate, collaborate, and take action online.

Developments in the agent ecosystem (AutoGPT, AI concierges, agent frameworks) are already enabling scenarios like this:

Bob wants socks. His AI assistant finds socks he likes, purchases them, and tracks the delivery - all using MCP exchanges between corporate and personal AI agents. The Boston Sock Company, meanwhile, runs on its own AI agents that manage design, supply chain, and customer support.

This isn’t sci-fi. It’s the early days of bots building bots, and they’re already changing how we work, research, and make decisions.

Fast-forward to 2028: The future of market research in an AI-enabled world

By 2028, the industry could look radically different:

• AI “concierges” will run lit reviews, collect data, and conduct analysis - often without human initiation.
• Research cycles will accelerate dramatically - some projects completed in near real-time.
• Innovation patterns will shift, with AI discovering patterns across disciplines that humans never considered.
• New roles will emerge: _AI liaisons, ethics moderators, human insight governors.

And as for traditional roles? They’ll shrink. But they won’t disappear. Instead, they’ll evolve.

Thoughtfulness in the age of AI: Ethics, judgement, and creativity

As AI becomes more convincing, there’s a danger of trusting answers that sound right but lack real-world grounding. As one writer put it: “The ideal solution would be for language models to refuse to answer when their predicted accuracy falls below a threshold. But that’s not great for a VC pitch deck.”

So, what matters now?

Not just smarts, but judgment, ethics, creativity, and interdisciplinary thinking. We’ll need researchers who can work across a society of minds - human and machine - and represent the richness of human experience in the questions we ask, and the answers we accept.

If the 2020s were about experimentation, the 2030s will be about coexistence.

Not man vs. machine. But something messier, more complex - and ultimately more powerful:

Man with machine. Mind with many minds.