In the age of AI, everyone is thinking, at least it appears. But are we really thinking, and what is the future of thinking?
I am not a big fan of the Slovenian philosopher Žižek. I find him too aggressive, yet in one of his interviews with the Institute of Arts and Ideas in London, he argued that AI is bypassing in-depth exploration and that users are increasingly reading, referring, and thinking through AI. Indeed, AI could be an amazing aide, yet is it not the case that we think, therefore we are?
I am not sure to what extent humans can think independently amidst the confident, structured communication and suggestions of different AI models. This brings me to Giddens’s structuration theory, where if we consider AI as structure, it could in the future be both the medium and the outcome of our individual and social action. In simple words, we shape AI, and AI shapes us. The relationship advice of ChatGPT, brainstorming with Claude, or even simple travel planning with Perplexity; we are, unlike with other technologies where we were simply users, now in no less than a direct co-constitutive relationship with AI.
This kind of human-technology relationship is not entirely new. Television and social media have been critical parts of our thinking and mediums of social change. Writers, activists, and thinkers using these mediums present their ideas to the general public and mobilise them to bring people together; some efforts may backfire and some may succeed. However, in all these cases, we could trace a person who could be rewarded or held accountable for those ideas. Yet in the case of AI, whom should we look to, whom do we blame or reward for the ideas that shape our thinking and probably doing?
This question becomes more important for populations that believe technology is sacrosanct and functional without any hidden interest. This is particularly true for the large section of smartphone users with underdeveloped critical thinking capabilities, in contexts where education systems and individual aspirations evolved around capitalist development following the industrial revolution and post-colonialism, and were meant to serve the needs of the product and labour markets. In such a world, human-AI interaction could be risky, because whatever AI suggests will shape their thinking.
And it is not only vulnerable populations who are affected. AI integration could also transform how all of us think irrespective of the country and contexts. Earlier, whenever we had doubts, questions, or needed clarification and advice, whether in personal life or at work, we sought out experienced people around us, visited libraries, or checked reliable sources on the internet. Now, circumstances are different. Despite having critical thinking capacities, many of us are preferring AI suggestions first, and when the response resonates with us, we are less likely to question further, perhaps because we are inherently lazy. In this process, however, we are becoming less connected to each other and more connected to AI models. This shift has implications for our thinking and for the social organisations through which we seek collective representation.
So, what can we do? AI is easy to access, it is a quick fix, and it is mostly correct — why not benefit from it, why not think through it? One may ask this genuine question. I believe that doubt in AI and confidence in oneself are both essential for a meaningful thinking process, yet neither comes easily without evidence. Therefore, observing, reading, failing, and repeating the process again and again whilst self-reflecting, now, indeed, with the help of AI, is, I believe, essential for the future of thinking in our society.
About the author: Dr. Adarsh Kumar is a researcher at Friedrich Schiller University Jena, Germany. His work focuses on algorithmic control, worker well-being, the future of work, labour economics, and the political economy of digitalisation.