AI Storm : The societal impact of AI

Anyone who has not been living under a rock for the past year, irrespective of whether they are remotely related to technology, will acknowledge that ChatGPT has changed the technology landscape massively, in a way it was never dreamt of, even by the founders, and investors of OpenAI ( the company behind ChatGPT ).

This type of massive and widespread impact ( it is estimated that ChatGPT attracted 1 million users in just one week ), and palpable interest or excitement with technology among the public has not been witnessed since the release of smartphones.

Truth be told, it is an immensely useful tool, be it for asking questions, and getting answers, to write a nice goodbye post to a colleague who is leaving, to creating complicated reports on how disaster recovery should happen in a cloud-agnostic ecosystem.

But how exactly will it shape the general conscience of the society ? Will it unite us as a global population, or will it divide and segregate us even further ? Will it inspire us or will the fallout of AI adoption result in an even more divisive and unequal society. In this article I look at some of these broad aspects.

Openness

At its core ChatGPT makes knowledge more accessible, which in theory leads to a more open society.

But, how does these Generative AI models ( that powers ChatGPT ) get the answers to all these questions ?

At a high-level, these GenAI models are LLMs ( Large Language Models ) that are trained on public text ( Wikipedia, Internet, the Web, etc ) and they can refer to the same when answering questions.

But the fact that the companies monetizing the insight from this data, have not found an effective way to ensure that the authors, or custodians of that data are fairly compensated, has led to an issue.

Content generators are thinking of restricting access to their data stores, in the fear that these LLMs will mine their data, without paying them royalty of any sorts. This has led to sites like Reddit, and Twitter exponentially increasing their data access API fees, in the hope that companies like OpenAI who mine that data for their models will now have to pay them a substantial part of their revenue if they want to provide answers based on that data.

In that sense, the same technology that was designed to promote openness and collaboration, in some cases, is indirectly encouraging data hoarding, which for society, might not be the best direction to move towards.

Quality of life

At its essence, it is widely recognized that AI as a whole, and GenAI tools like ChatGPT will considerably improve the quality of life, because the mundane, repetitive work will now be automated or delegated to these GenAI bots, and we as humans will be able to usher in an era of productivity, and creativity.

This is true, but as we have seen with earlier technology revolutions, the short-term impact will be that a part of the workforce will become redundant. Yes, there will be new jobs focused on GenAI tools, and processes like Prompt Engineering, etc, but do we honestly believe that the workforce that will become redundant due to AI, will be able to upskill themselves enough to learn Prompt Engineering, AI modelling, Alignment, etc and secure these new set of jobs that AI will bring.

What about the clerk in the bank store, whose job got redundant due to a new ChatGPT based interface that does the same work; will he be able to learn new AI lingo and become a prompt engineer ? Quite unlikely.

However, this is not something specific to GenAI, or ChatGPT; with any significant technical change, this issue will be omnipresent, and we as a society, need to do better, become better version of ourselves to be able to take all strata of the society along with us in this incredible technology upliftment journey.

Social Impact : Bonding

Concerns have been raised by different segments of society regarding potential bias or tailored responses provided by ChatGPT. While debatable, if doubts persist, society as a whole will struggle to fully embrace the power of this technology revolution.

Big Tech companies driving these LLMs must continue promoting best practices to ensure fairness and openness. Additionally, they must take further steps to instil public trust in the content generated by their applications.

A big part of the population will leverage the output of these LLMs as-is, and maybe even live a part of their life, or take a critical decision based on those answers. As the producer of the answer, Big Tech ( OpenAI for example ) needs to be more accountable for the content their applications generate. Mere disclaimers advising users to mistrust or manually verify outputs may not be an acceptable solution.

As a society, we need to come to some sort of a shared understanding of what is true, of what is fair, and then inject that understanding into these LLM(s) so that they act as an extension of that similar societal conscience, and not exists as a standalone AI model inside a vacuum.

Conclusion :

I truly believe that the impact that GenAI can have on making a better version of our society is massive.

I do not recommend moving away for this path of experimentation, of introducing new features that showcase how far, we as mankind, have come in the evolution of technology, but the final outcome of any transformative technology cannot just be productivity, or automation.

With this sort of technology, we truly have a chance to shape, influence, and uplift society, for the common good. We need to acknowledge the risks, and shortcomings with regards to the overall societal conscience, and then go back to the drawing board to ensure that we can do better.

I hope we, as humans, do the right thing !