Chat GPT and the ‘epistemic economy’

The reasons why some technology such as Chat GPT catch fire whilst others fall by the wayside means shifting our focus from the tech itself to the users

ChatGPT and the other versions of generative AI have upended technology markets in ways that seem to have taken many commentators by surprise. The media has been full of discussion of the possibilities, people have been talking to each other, trying it out, sharing what they found.  But why has this gained so much excitement when earlier versions did not seem to make all that much of an impression on the population? 

This is a question that lies at the heart of much head scratching when it comes to technology innovation: history is littered with examples of tech development that seem to often amazing leaps forwards but do not enjoy wider adoption.  And there are other cases where there is widespread adoption of something that involves very minor technology development.

Some have tried to create a science of technology evolution to explain this, placing the onus on the technology itself and the way in which has its own ‘natural’ patterns of progress. It can seem that this is the intuitive way we answer the question of technology progress: we start with the technology and then ask what aspects of it we like or find useful or are excited by.

The benefit of the time we are in right now is that we have an opportunity to see close-up what is happening and start to unpick the dynamics of this reaction to technology change. With this in mind, we suggest that there are benefits of turning 180 degrees from the technology itself, to look more closely at the other part of the equation: the user. 

Innovation from users (not just the tech)

Of course, staying close to the needs of users to drive innovation is not something especially new, there has long been enthusiasm for engaging people to co-create products and services reflects this orientation, with Lego as one example of a brand that successfully makes use of user-generated content. On the Lego Ideas platform, people are encouraged to share ideas for new sets. These ideas are voted on and the best ones can become official sets. In this way user-generated content can lead to a user-generated products. 

While this has much to recommend it, tapping into user-generated content can also be fraught with challenges, not least given the huge amount of content that is often created. Another way that users can be engaged to support innovation and the successful adoption of new products or services is through Lead-User Theory. This can help to identify users whose innovations are likely to succeed. These users are defined as those that 1) can see the benefits of a solution to meet their needs and 2) are ahead on an important trend in the marketplace.

Applying this thinking to the excitement around ChatGPT, we can see how the provision of open access encouraged engagement by Lead Users, whose reaction has made clear they perceive their needs are being met in a very tangible way by this solution.

There is a lesson here surely for organisations seeking to innovate – encourage Lead Users to engage early on and assess the degree to which there is take-up and engagement (or not). But this does begs the question of how these needs were created in the first place. 

The epistemic economy

That we have a drive to collectively orient ourselves to make sense of the world is something we have covered before:  we regularly cite the work of Sloman and Fernbach as well as Nick Chater and George Lowenstein with a recent paper that centres around this theme. But the way we are collectively responding to ChatGPT perhaps shows us more clearly and saliently how this ‘under-appreciated’ drive is operating.

We live in a world where it is often considered that we have a huge challenge in how we make sense of it. For example, the WHO set out how we are living in an infodemic, overwhelmed by huge amounts of information as a result of an ever more complex and confusing world. Hence, there is a need to try and navigate this ‘epistemic economy’ effectively, to make sense of and tame this complexity. 

We can see how we the notion of living in an infodemic shapes how we define our needs for sense-making.  There is plenty of evidence to indicate that concerns about information overload is a burden on now we feel with, for example, 73% of people wishing they could slow down the pace of their lives, 83% considering the world is changing too fast, underpinned by the way many people feel that technological progress is destroying their lives (61%)

We are familiar with the way we can locate knowledge in all sorts of online locations, from a more familiar web search yielding pages of links, online forums with lively discussions, to email newsletters and social media posts.  The salience of this promotes a sense of an epistemic environment that is unwieldy and out of control despite the fact that our local bookshop or library has long had more material on the topics we want to explore than we could ever likely assimilate.  But regardless of whether or not we really are in an epoch of information overload, people simply feel the anxiety of this more keenly and are ready for a means to fix it. 

From this we can then piece together how the ‘epistemic economy’ has reached a state of readiness for ways to address our collective concerns, to make our lives simpler, to offer a means to synthesise and make sense of all these different information strands without us having to do it all ourselves.

The degree to which current generative AI solutions offer solutions that are robust, accurate and representative of the range of evidence that is available is something that is the source of a huge amount of discussion.  We are limiting ourselves to an analysis of why there is so much discussion and engagement and this, we argue, points to the importance of growing concerns about the ‘epistemic economy’ shaping what we prioritise and how we react.

In conclusion

People such as political theorist Paolo Gerbaudo speculate that we are witnessing a moment of global transition of ideas, aligning with historical cycles in ideologies that take place every fifty years or so.   Kenneth Gergen  argues that social sciences need to reflect these shifts, not least because what was shaping behaviour decades ago is not what is shaping behaviour today.

We are collectively sense making not just of the information itself (ontological sense making) but also the mechanics of how we go about doing this (epistemological sense making).  We are collectively changing our orientations towards the knowledge tools we have, moving away from some (there is, for example, a decline in book reading, as well as newspapers)  as connecting us to others (such as the growth in online forums such as Reddit and messaging apps such as WhatsApp and Telegram). Behavioural science then is as much about unpacking and understanding culture and the mechanisms that underpin this collective sense-making activity, as it is looking at what is going on inside individual heads. 

Engaging people in the innovation process is key, through co-creation and Lead-User participation. But we also need to think more deeply about the wider cultural mechanisms that are shaping our needs and expectations of these research participants.