As AI race heats up, let’s pause and ask: "ChatGPT is the answer to what problem? "
“A race starts today,” declared Microsoft CEO Satya Nadella as he launched Bing search engine powered by the artificial intelligence within ChatGPT, putting Google executives on edge.
In response, Google quickly announced it will revamps its products with the debut of Chatbot Bard, but a single slip-up from Bard cost the company a staggering $100 billion. This AI arms race comes hot on the heels of recent layoff announcements from big tech companies.
With investors and tech pioneers buzzing with anticipation over the potential of the new technology, it’s crucial that we ask about the original objectives that sparked its creation. Is the innovation simply born from pure “curiosity”, as Mira Murati, OpenAI’s Chief Technology Officer suggests? Or is it driven by a greater purpose and vision? And if so, do these aspirations align with society’s pressing needs?
More than product
ChatGPT has achieved a major milestone by reaching 100 million users in just two months, fueling widespread excitement about its potential to revolutionize various sectors, from customer service to education and entertainment. It has already made its mark on Bing and could prompt Google to react by updating its search engines with its experimental AI technology— raising serious questions about the transparency, source verification, and user interactions in search systems.
ChatGPT (and “generative AI” generally) aims to close the gap between human language and machine understanding by making communicating with computers easier and more efficient. Don’t forget we are talking about just English here.
They are designed to generate human-like responses to a human prompt, much like psychotherapist chatbot, Eliza, which was created 57 years ago by Joseph Weizenbaum to make conversation with machines possible.
The tech has already sparked both excitement and fear among the masses and institutions alike. Now, instead of struggling with an overwhelming number of links and pages, ChatGPT digests vast amounts of information available online and serves us a version of reality we want. Microsoft CEO Satya Nadella envisions a future in which ChatGPT “reshapes pretty much every software category that we know.”
This is not the whole story though. ChatGPT, like any other innovation, is not just a product of tech enthusiasts, investors, and end-users. It also emerges from and impacts on the larger ecosystem of stakeholders.
For ChatGPT it includes underpaid Kenyan workers, and the environment that must bear the brunt of its energy-intensive technology. This justifies the actions of governments and regulatory bodies to protect society from its undesirable consequences.
A Solution looking for potentials, not problems
From drug discovery to content creation, code writing, and essay writing, large language models can now be utilized in a variety of potential areas. Mira Murati, OpenAI’s Chief Technology Officer emphasizes on the tech’s “immense potential to help with personalized education.” She went on to say, “We weren’t anticipating this level of excitement from putting our child in the world. I’m curious to see the areas where it’ll start generating utility for people.”
AI is always been understood as a solution looking for complex decision-making problems from car driving to financial portfolio management and healthcare diagnosis.
AI innovators build a product and then think to themselves, “Who would want that?”, or “how could it generate utility for people?”
Of course, the world of science and technology is replete with examples of serendipitous moments, such as the discovery of Viagra, when innovators create a failed solution for chest pain and then find out it’s a solution to the problem of erectile dysfunction. A product that made tens of billions in revenues for its owner, Pfizer.
Microsoft’s recent launch of a new version of Bing search engine powered by ChatGPT is a testament to how AI rapidly can move from curiosity to a revenue generator.
However, we should face the fact that the story of ChatGPT is not serendipitous as the development of Chatbots has been underway since the creation of Eliza. Companies like OpenAI have received ample funding from prominent investors, and those who have their hands on the market pulse.
Unlike the development of the Covid-19 vaccine, new technologies often lack a societal and environmental focus. Instead, the approach is often to build a solution first and then search for potentials, turning them into legitimate concerns or problems.
With the United Nations’ Sustainable Development Goals serving as a global platform for action, it is crucial that innovators create solutions that help us tackling world’s most pressing challenges, including poverty, inequality, climate change, and environmental degradation, not those can exacerbate them, reinforcing structural injustices perpetuating systemic problems.
I am sure a new group of AI practitioners will soon make the argument that generative AI are here to stay and must be leveraged for the public good.
Of course, we must explore how to use these tools for the benefit of society. But the most fundamental questions remain unanswered:
what problem is ChatGPT actually solving? What’s the business model behind it?
We build a solution first and then search for potentials, turning them into legitimate concerns or problems.
Resisting napkin technology
With big tech companies now carelessly rushing to launch their state-of-the-art products, it’s important for us to pause and reflect on the direction we are heading in.
Our preoccupation with the feasibility of AI development, rather than its purpose, threatens to imprison our social imagination.
We must strive for alternatives that better address society’s genuine needs and problems.
Inclusion is a key tenet of responsible innovation, but it is equally important to consider when, where, and how we involve diverse voices and ask difficult questions. Otherwise, responsible innovation risks becoming little more than a marketing campaign to portray a positive image.
Isn’t it too late for OpenAI just now seek input from social scientists and experts in humanities in the discussion about ChatGPT, when the tech has already been imagined, developed, and is on the brink of becoming a part of our daily lives? And if not, will the developers be open to hearing critical perspectives that question the rationale behind ChatGPT?
Investors must also align their investments with society’s broader sustainable objectives.
There’s always a danger of our limited resources being squandered on unnecessarily complex solutions, much like ‘Self-Operating Napkin’ technology in Rube Goldberg’s famous cartoon—where a man uses a complicated machine to clean his chin while he’s busy chowing down.
It’s a cautionary tale for a misplaced focus of technology.
Ehsan Nabavi is a Senior Lecturer at Australian National Centre for the Public Awareness of Scienceand the Head of Responsible Innovation Lab at the Australian National University.