Word of the Year 2026? Be WEIRD…It Is Your no.1 Asset

“Based on my using LLMs for a decade…In an AI driven world NOTHING matters more than #originalvoice” ***

I have been using that line in commenting on social media, part taking in think tanks and discussion, responding to interviews and advising clients on the constant tsunami of AI commentary for near 3 years. The more the world uses AI or LLMs or any digital built models for collecting, analyzing, commenting, summarizing, creating the more your job is to be different. 

Most of us will never develop LLMs. We will use those available one or copy them. You may decide you like Canvio’s newest edition, or i-Genie as a wonderful partner, or prefer Perplexity to Claude or Deepseek, or use Mistral because “well it’s French so it understands us better” as one Parisian colleague told me last week. 

I am on a number of forums that discuss the good and bad of all kinds of platforms that can help with near all aspects of the market research business. Just 2 weeks ago Junu Yang** wrote a great piece on this blog about how using AI for qual is changing and being changed and will change. Learning to use AI can be fairly easy or a nightmare depending on the level you want to go to. But using it is not the point…

USING is the start…using it in a WEIRD way means your worth something.

In that Junu Yang article he summarizes many things that my mate Craig Griffen has been talking about in his Campaign for Real Qual*** the last couple of years : AI is here, it will help, learn to use it … but learn to use it in a way that integrates the tech and human. 

Use It...But Integrate It in to You to Make Your Self Stand Out

Over the last year or so I have been asked to speak at around 15 conferences and events for all kinds of industries including 4 large events for the market research industry. I started all of them by asking the audience to undertake a simple exercise :

“please take out your phone of tablet of laptop, go and open your usual platform of preference .. GPT, Claude, Perplexity, Mistral etc … and enter the following prompt : tell me what makes ( enter your name or your companies name ) different or unique”

In other words do a very quick and basic brand survey on yourself. I then ask how many people in the audience have done that before, how many regularly search themselves and or their brand on google or other search platforms, or AI based platforms. Shockingly to me the biggest response has been the looks ( the actual facial expressions ) in the audience. The slight shock, the slight flush of embarrassment, the shaking of heads. For example in the 4 MR industry events where I addressed maybe 1,000 people no more than 50 admitted ever doing that kind of search.

In other words an industry of researchers who never research themselves to see if they are unique.

When I dug deeper with some audiences almost none had actually ever used a 3rd party platform to track themselves or their brand. And almost no one had ever heard of hiring an independent agency to research and or track their brand … or themselves.

Now an open admission … I do. I have done a google search of both myself and my small consultancy, Bibliosexual, at least once a month for a decade. When I started using LLM platforms in 2015 one of the first things I did was track my personal brand globally and in key markets I mostly work in at least quarterly. Looking for what potential clients, partners, industry members would find out about me. Tracking, adjusting the content by reacting to it, pointing out errors or things missing .. yes manipulating the platforms in the same way we would hope to help clients to use research to adjust the way their world sees them and to make my/their brand unique, weird.

Vanity ??

No just good business practice.

And in a weird world ? ESSENTIAL.

Now we can leave the whole “why are market researchers so pathetic at getting themselves researched” question for another time. What this exercise really was about was helping audiences understand that in a world where LLMs have become the default “investigation mode” for most of the world you need to make sure people searching you not only find you but find out why you are different, original … weird.

The 3 “Ways of the Weird” You Need to Work on:

First make it a habit to be weird not WEIRD

I would hope you have all heard WEIRD ( Wester English Industrialised Rich Democratic ) and weirdness used as shorthand for what AI misses or flattens out. The non-average, surprising, context-heavy aspects of human behavior that make brands interesting and insight rich. These words signal both a critique of over reliance and as reminders that humans can do better or add real difference. 

AI models are trained to find patterns and predict the “most likely” next word, image or outcome. That tends to pull results toward the familiar and average. So results may be “correct” to a certain point but can feel sterile and ordinary.

But let’s be honest rough, inconsistent, idiosyncratic content feels more human than polished, templated AI output. 

So for example, you may note “typos” in my article here, or in fact in any report or article or post of mine. Because I prefer english english, refuse to use spell check, refuse to use grammarly or any other “tool” which takes away my uniqueness…or as the word of the year implies, my “weirdness”. Despite 30+ years living in Asia and working on projects globally I still talk and sometimes write with an Aussie accent and try to write in a slightly annoying style. All of which helps make me weird in some way…and therefore noticed and remembered. 

WEIRD bias has been discussed since data bases began. I can remember a discussion about the WEIRD bias of LEXUS NEXUS back in the 1980s ( yes I am that old ). In the last 3 years we have seen a growing number of articles and commentaries pointing out how the major GPT/LLM platforms are weird in the bias of their data base and analysis. That will change as they add more content from non Western, non English sources…we hope? No, we know. 

There have been LLMs around for near 20 years that avoided that for of WEIRD. Platforms like SignificnceSystem allow you to use over 40 languages to investigate the whole of the internet in any country where they are used. Want to track trends in France or how France sees them versus the UK or Japan. No problem. Yes you can use Mistral or you can use those older platforms to get a non-weird, French view of any subject. 

Then develop a culture of understanding and delivering the weird

But this weird thing becomes even more relevant when we consider context as the key to research process and analysis. 

Lessons learned in that 30 years experience working in Asia. The importance of things like silence ( in many cultures the silence IS the really import finding ) or context ( such as the Japanese cultural habit of interpreting how to answer a question based on the background around the questioner ) or body language. My early mentor in qual research 4 decades ago taught me that you never interview people over a table because you need to see their feet to understand how comfortable they are with the discussion.

All of this adds to the “weird” not WEIRD that is so essential to good market research. All sorts of platforms are working on these shortfalls. But based on earlier tech that promised researchers would not be needed ( think the spreadsheet, internet, the original idea of “social media” not as a gossip tool but as a sort of internet driven, extrapolated Wikipedia, etc etc ). Either the tech did not grow as thought or more likely researchers individualized, integrated in unexpected ways and stayed weird in their use of the tool. 

So look for techniques that adds to the use of any tool in ways no one else does, then present your findings weirdly. No you don’t have to misspell. Find new ways to speed or slow or combine techniques. If every one else is using AI generated charts then massively manually change to something weird, look at how others deliver findings and what findings they get too and make sure any client says “ well yours was different”

Because weird is what drives demand

One conference I did that little trick with was a gathering of HR people. They may not have searched themselves much but they told me doing it to candidates was now normal. They don’t read resumes or cover letters, they don’t google you or check your FB, LI, TikTok. That is sooo pre-covid. No. They ask popular LLMs what makes you interesting, unique, different…in other words why are you weird in what you do. 

Because no one wants a me too brand or product. We researchers should know that of course. After all in 40 years developing brand stories, marketing campaigns and researching what to do did anyone ever ask me to “make me look the same as the opposition”. 

This is not me being weird. This is a market reality. Earlier this year articles in Forbes and other places started appearing with headlines like “Succeed on Linkedin in 2025 by being more weird”. It should not be news.

Whenever faced by a tech challenge the winners use it differently, create uniqueness with it or against it, look for ways to be different.

Worried about AI in 2026 ? You should be. IF you are not being weird. 

* https://www.greenbook.org/insights/focus-on-apac/original-voice-the-most-important-goal-for-market-researchers

** https://www.greenbook.org/insights/focus-on-apac/ai-wont-replace-qual-researchers-but-it-will-change-everything-about-the-job

*** https://www.greenbook.org/insights/focus-on-apac/the-campaign-for-real-qual-four-provocations-in-an-ai-driven-era