Should we be afraid of the new AI?

Is artificial intelligence (AI) finally coming of age? If computers start reasoning and working more like humans, what should we be excited—and worried—about?

The world is currently agog trying out ChatGPT, the newest kid on the block. ChatGPT is generative AI: it isn’t intelligent as such, but it has a huge trove of data that it can access, which it uses to generate very human-like answers to quite complex questions. ChatGPT is open to all at no charge (for now). It has reportedly clocked up more than 100 million users already. I have been giving it a whirl for the past two months, and the results can be fascinating.

For example, I asked it to answer the question I put in the first paragraph above. Specifically, I asked it what Sunny Bindra would say about about the future of AI. The answers were captivating!

The first thing Sunny may say (said ChatGPT), is that “AI will continue to infiltrate our lives and work in ways we cannot yet imagine. As AI systems become more sophisticated and capable, they will increasingly be used to perform tasks that were previously thought to be the exclusive domain of humans.”

Next, my bot friend said that I “may point out that AI is already causing significant disruptions in many industries, and this trend is likely to continue. Companies that fail to adapt to the new reality of AI may be left behind.”

It also suggested I would be concerned about ethical concerns: “Bindra may argue that as AI becomes more powerful and ubiquitous, the ethical issues surrounding its use will become more pressing. These may include questions about privacy, bias, and job displacement.”

ChatGPT picked up a more nuanced point, one I often make on this page: “Collaboration between humans and machines will be key: Bindra may argue that the most successful applications will involve collaboration between humans and machines. As AI takes on more routine tasks, humans will be freed up to focus on higher-level tasks that require creativity and critical thinking.”

And finally, it warned: “Bindra may argue that the pace of technological change, including the development of AI, will only continue to accelerate. As a result, individuals and organizations will need to be highly adaptable and agile in order to thrive in the new reality.”

Ahem. I look at that and I think: yep, I would probably have said those exact things if asked, in quite similar language. Which begs the question: does the Sunday Nation editor still need me? Can’t he just dump me and use a bot masquerading as me?

Troubled, I asked ChatGPT a follow-up question: could you just replace Sunny Bindra? It told me this: “As an AI language model, I can provide information and insights based on the data I have been trained on, but I cannot completely replace Sunny Bindra or any other human consultant, speaker, or writer. This is because there are certain skills and qualities that are unique to humans, such as empathy, creativity, and intuition, that cannot be fully replicated by AI.”

Phew! It went on: 

“While AI can assist in many areas, such as data analysis and decision-making, there are still many aspects of consulting, speaking, and writing that require a human touch. For example, a human consultant like Sunny Bindra can use their experience, judgment, and interpersonal skills to understand the nuances of a particular situation and offer personalized advice that takes into account the needs and goals of their clients.

Similarly, a human speaker like Sunny Bindra can use their charisma, energy, and storytelling ability to engage and inspire their audience in ways that AI cannot. And a human writer like Sunny Bindra can use their unique voice, style, and perspective to craft compelling narratives that resonate with their readers.

In summary, while AI can augment and enhance the work of human consultants, speakers, and writers, it cannot completely replace them.”

Whilst I was reassured by this answer, I also noted the nuances contained therein. ChatGPT knows where humans excel: the human touch. Humans have interpersonal skills; they use charisma; they can judge other humans in special ways and understand the emotions underlying their behaviour.

To me—and, it seems, to the AI—the way forward is very clear: we should not run away from AI, we should not suppress it, and we should not fear it. If used properly, it will augment and supplement the capability of humans. Just as the calculator and then the spreadsheet freed us from the time-consuming drudgery of calculating numerical results repeatedly, AI might do our basic research and thinking and reporting for us, leaving us free to focus on creativity, nuance, empathy, and intuition.

Or wait: is that what it wants us to think? Is it trying to lull us into a false sense of complacency about AI, so that we allow it grow unfettered? I will be reporting on my continuing adventures with AI on this page. But if I disappear suddenly, be very afraid…

(Sunday Nation, 5 March 2023)

Like it? Hate it? Engage here