AI at Work: How We’re Actually Using It (Not How the Headlines Say We Are)

If you read enough headlines, you’d think artificial intelligence is either going to save civilization or end it.

In my experience so far, the truth is far less dramatic – and far more useful.

Like most people who help run businesses, I didn’t wake up one morning thinking, Today I shall revolutionize my workflow with large language models. My introduction to AI was more practical. A colleague suggested I try it for editing something I had written. I did. It was helpful. Then I tried it for something else. Then something else.

Before long, AI had quietly become part of the daily rhythm of how I work.

Not instead of people. Not instead of judgment. But alongside both.

From where I sit – in a research-based brand consultancy – the use of AI really shows up at three levels. Each builds on the one before it: personal productivity, team productivity, and what I’d call insights optimization.

Let me explain what that looks like in practice.

Level One: Personal Productivity

I’ll start with a confession: I like writing, but I don’t like rewriting.

Unfortunately, much of professional writing is rewriting.

AI has become a helpful sparring partner for that process. I’ll draft something – an email, a proposal section, a blog post like this one – and ask the tool to tighten it, challenge it, or reorganize it. Sometimes the edits are good. Sometimes they’re terrible. But the conversation itself sharpens the result. Sometimes, I work it in reverse – and AI helps me by providing a starting point.

In that sense, AI behaves less like an author and more like an extremely fast junior editor who never gets tired.

Beyond writing, I use AI to help synthesize secondary sources. When I’m preparing for a client meeting in an unfamiliar category, I’ll gather articles, reports, earnings transcripts, and other materials and ask the system to identify key themes or summarize what competitors are saying.

It’s not perfect – and I always verify anything that matters – but it’s an extraordinarily efficient way to orient yourself quickly.

The same applies to competitive awareness. Our clients operate in fast-moving markets, and keeping track of what competitors are launching, saying, or signaling can be time-consuming. AI can help aggregate that information and highlight patterns that might otherwise get lost in the noise.

Beyond work tasks, it helps me diagnose that ‘check engine’ light, respond intelligently to an insurance proposal, find the next book I should read. It simplifies life’s peripherals so I’m better able to navigate things.

In short, at the personal level, AI acts as a force multiplier. It doesn’t replace thinking – it gives me more time to do it. It doesn’t take action, but helps prepare me to.

Which brings us to the next level.

Level Two: Team Productivity in Research

Our firm specializes in insights, which means we deal with large volumes of information – qualitative and quantitative data that we capture, manage, and filter. While it used to be that data was hard to come by, that’s no longer the case. We’re drowning in it – from organized, large data sets to social media – and the trick now is how to make sense of it.

Historically, some of the most valuable insights in research came from open-ended responses: the written or spoken comments people provide in surveys, interviews, and focus groups. The challenge has always been scale. Humans are very good at interpreting nuance, but reading thousands of responses takes time.

AI can help accelerate that process.

For example, we now use AI-assisted tools to help organize and code open-ended responses. The system can cluster similar comments, identify themes, and highlight unusual or emerging ideas. That doesn’t eliminate the role of researchers – in fact, it arguably makes our role more important. And we get to spend less time on administrative tasks, more on the crucial sense-making that drives impact.

Why? Because interpretation still matters.

AI can tell you that 18 percent of respondents mentioned “ease of use.” It can group comments together that appear related. But it can’t determine what that theme means for a brand’s positioning, product design, or growth strategy. That’s where experienced human researchers come in.

In our work, AI helps us move faster through the mechanical parts of preparation or analysis so we can spend more time on the thinking parts – the interpretation, the storytelling, the implications.

We’re also deploying AI in the design phase of research.

It can help pressure-test survey questions, suggest alternative ways to phrase concepts, or propose exploratory angles we might not have initially considered. Sometimes those suggestions are insightful. Sometimes they’re off the mark. But again, the value often comes from the dialogue.

AI can also play a role in data collection – we’re experimenting with AI conversations, in which consumers dialogue with AI the same way we all do in our personal lives. The benefit of that is scale – if AI can conduct, let’s say, 100 rudimentary consumer in-depth interviews in the time it takes us to conduct 1, there are major efficiency benefits.

Think of AI less as outsourcing research and more as expanding the toolkit researchers have available to them. The human role – judgment, curiosity, skepticism – remains central.

And that leads to the third level, where things become even more interesting.

Level Three: Insights Optimization

One of the longstanding challenges in the insights world is that organizations accumulate enormous amounts of knowledge – research studies, brand trackers, customer feedback, social comments, syndicated data, eCommerce purchases – but struggle to fully activate it.

Important findings get buried in slide decks. Institutional knowledge lives in people’s heads or vast and impenetrable SharePoint folders. Teams repeat studies because they can’t easily access what has already been learned. Important signals fall into the cracks between disconnected data sources.

This challenge is magnified in an era of accelerating innovation and a real-time need for decision support.

This is where AI can play a transformational role.

At Finch Brands, we’ve built an AI-powered platform called Charlie by Finch Brands™ to help address this challenge. Charlie blends advanced knowledge management with AI-driven synthesis across multiple sources: past research, social intelligence, customer feedback, and other inputs.

The goal isn’t to replace research. It’s to make research – and insights more broadly – far more usable, connected, and accessible.

Instead of combing through years of reports, teams can ask Charlie questions and surface relevant findings quickly. Patterns across studies can be identified. Signals from social conversations can be connected with formal research results. Insights that once sat in separate silos can be brought together.

In many cases, we’re using Charlie alongside clients – either to help them explore existing knowledge or to integrate new research findings into a broader intelligence ecosystem.

The result is that insights teams are faster with answers, in better command of resources, and in a stronger position to make the desired impact.

Researchers, strategists, and marketers still ask the questions, frame the problems, and interpret the answers. But the tools now help them navigate complexity faster and with greater perspective.

Humans Still Run the Show

Whenever a new technology emerges, there’s a temptation to frame it as a replacement for people.

In my experience so far, AI works better as an amplifier.

• It amplifies productivity by handling certain mechanical tasks faster.
• It amplifies curiosity by enabling quicker exploration of information.
• And it amplifies human expertise by enabling professionals to spend more time on judgment and less time on logistics.

But it does not replace the human ingredients that make insight valuable: empathy, skepticism, creativity, and experience. Those things still come from people.

The companies that get the most from AI will be the ones that remember that – and who get better at using AI to promote individual, team, and organizational productivity.

For our team, the approach has been simple: experiment thoughtfully, stay grounded in real client needs, and treat AI as a tool rather than a threat or a miracle.

So far, that’s working pretty well.