Artificial Intelligence (AI) is reshaping the insights ecosystem, not by replacing human intelligence (HI), but by amplifying it. Nowhere is this more evident than in the evolution of online communities, which are becoming powerful engines for customer-centric decision-making. When AI and HI work together, brands can unlock deeper insights, faster decisions, and broader engagement. Thereby keeping the voice of the customer at the heart of the organisation.

In this post, we explore and share how blending AI and HI can supercharge your community strategy and drive real business impact.

Understanding the Roles – What AI does best and where HI Leads

In order to leverage AI, we need to understand what it is good at, where it needs to be guided by humans, and where humans should provide the key elements.

AI Strengths:

• Ability to access large amounts of data.
• Ability to do the same thing, again and again, at scale.
• Ability to re-arrange known ideas into an enormous array of new alternatives.

AI Weaknesses:

• Knowing which problem to tackle.
• Understanding what works with the end client and what works with the community.
• Setting priorities.
• Deciding which ideas to use.

Where AI Adds Value to Research

In this section we outline the key ways that AI is being used today for wins in the context of insights and online communities

Project Design

[e.g. briefs, proposals, questionnaires, discussion guides, house styles, stimuli]

At the start of a project, routinely use Deep Research to find out what is already known. Give it a description of what you are planning to do and ask it to search for what is already out there on the web. It will return with a summarised report, with links to its sources.

If you have received a brief or a proposal etc. ask the AI to summarise it. Ask it what the key points are, what the must-haves and nice-to-haves are, or what might be left unsaid. This will usually throw additional light.

When you are creating briefs, proposals, questionnaires or discussion guides, ask the AI to utilise a House Style document to speed things up and to make sure they conform to the way your organisation works.

A House Style can be created by asking AI to summarise your last couple of years’ worth of projects, and then you edit the document it produces. The House Style can be as simple as a PDF that you upload when creating documents, or you can create it as an Agent (e.g. Gem in Gemini, Custom GPT in ChatGPT, or Agent in Copilot).

Analysis

At the analysis stage, a big advantage of AI is its ability to work with larger amounts of text. If you have open-ended comments, then you can use Text Analytics, such as the one available with Platform One’s community platform. For things like discussions, you can upload them to the AI and ask it to summarise them, or even to conduct analyses such as content analysis or narrative analysis for you. (You can also ask the AI to explain content or narrative analysis to you.)

Increasingly, the AI platforms are adding in code interpreters to analyse quant data, for example ChatGPT and Copilot both run Python in an embedded interactive environment. This greatly expands the analyses you can run. Anthropic’s Claude uses JavaScript and graphics libraries to help visualise data in new and exciting ways.

Reporting

At the moment, AI is not great at producing finished reports, but it can help you produce reports. For example:

• By asking the AI to generate a first draft of the text you want to use (leveraging a Community Reporting Style Guide to make sure that key points are always made).
• By reviewing your reports. One great prompt is “What are the main points this presentation is making?” If the AI’s take is different from the one you intended, you have more work to do.
• By creating interactive deliverables. Within your organisation you can create an Agent that will allow your colleagues to talk to the data and to your analysis. Or, you could leverage Notebook LM internally and share it with colleagues.

Workflow

Clients have found the 10:20:70 rule to be essential to the successful use of AI by organizations:

• 10% is the quality of the AI,
• 20% is the quality of the data (in this case that includes the quality of the community), and
• 70% is the quality of the way the AI fits into the organisation’s workflow.

You need to be able to upload your existing file types, you need to be able to share working and finished documents with colleagues, and you need to be able to export outputs to your existing structures. This is one of the reasons that companies who use Microsoft Office favour Copilot and why companies who use Gmail and the Google tools prefer Gemini. Creating House Style documents, that are already loaded into the AI (for example, as Agents) is a great way to codify and standardise your projects, allowing you to democratise the process of using the community more widely and safely.

Key Steps

Here are five steps you can take to leverage AI and blend it with Human Intelligence to enhance your efficiency and effectiveness.

1. Create and maintain House Style guides that AI tools can reference for consistency, while human intelligence applies nuance and contextual judgment.
2. Use AI-powered text analytics to surface insights from open-ended data, with human interpretation guiding meaning and action.
3. Leverage AI for first-pass reviews of briefs, proposals, and reports, with human intelligence making final decisions based on context and strategy.
4. Apply AI for broad and deep research, letting human intelligence define questions, validate sources, and draw meaningful conclusions.
5. Co-create with AI for ideation and stimulus, then refine outputs through human storytelling, ethics, and audience awareness.

The Golden Rule: AI Should Empower HI

AI is a tool – not a replacement. The real power lies in enabling human intelligence to focus on what matters most: asking the right questions, interpreting meaning, and making strategic decisions. When AI supports HI, and communities are leveraged effectively, organisations can move faster, think deeper, and act smarter.