Ever since ChatGPT was made available to the public by OpenAI, it has significantly contributed to the advancement of ‘people accepting AI’, now that it can do so much within a short period of time.
Prior to that, we had AI doing so much in the background, doing the boilerplate work that didn’t require human intervention. Over the years, AI has always been developed and used in various real-life scenarios.
For example, your smartphone has an AI camera that can react to its environment without the help of the human mind. The camera software exhibits different cognitive functions like thinking, learning, and problem-solving, which are generally associated with the human mind.
Still, I would argue that despite these significant advancements, there are still key aspects of human cognition that make it unlikely for AI to completely replace humans.
Emotional Intelligence in AI would be extremely difficult to implement
I am confident that someone somewhere is working on making the AI systems emotionally intelligent. It’s just not there yet.
By definition, emotional intelligence (EI) is the ability to understand, manage, and use emotions effectively in interpersonal relationships. This involves empathy, self-awareness, motivation, and social skills.
Unlike AI, humans are capable of processing and responding to emotions, which are essential for building trust, collaboration, and effective communication. Yes, there might be people with less EI, but they still have some emotions that AI systems lack currently.
There are reasons why it would be difficult to replicate these factors for an AI.
→ Human emotions are highly nuanced and context-dependent, making it difficult for AI to accurately interpret and respond to them.
→ Emotional expressions and interpretations can vary greatly across cultures, further complicating AI’s ability to navigate emotional situations.
→ Humans can learn and grow emotionally over time, adjusting their emotional responses based on new experiences and understanding. AI, in contrast, would require constant updates and training to achieve a similar level of adaptability.
Creativity of AI is still not comparable
Some of you might not agree with me on this, but I am going to illustrate how.
The one thing that made the chatbots based on LLMs, such as ChatGPT, popular was their creativity. The ability to produce original, valuable ideas or solutions, often by combining existing concepts in novel ways, is creativity by definition.
However, DALL.E2 and Midjourney, the text-to-image AI image generators, have been accused of stealing art from artists. Imagine artists who spent hours and days putting out their thoughts and drawing their ideas on paper or a tablet.
Although there’s no denying that AI can analyze new ideas or solutions by analyzing vast amounts of data, one of my colleagues refers to these LLMs as “glorified datasets,” since all they do is train using gigabytes of datasets.
Human creativity is fueled by imagination, allowing us to see future scenarios, explore hypothetical situations, and create entirely new worlds in our minds.
That’s how artists are able to create impeccable artwork. That’s how writers are able to write stories that exceed human imagination. That’s how some programmers are able to solve complex issues that almost seem impossible to resolve. That’s how prolific filmmakers are able to make movies that are aspirational and inspirational. It is for their creativity, imagination, and skill that they are paid highly.
As of this writing, I have not seen an AI that can imagine and create like a human. Maybe future language models will be able to do so.
Overall, I believe AI will continue to play an increasingly important role in our lives. It will definitely replace a lot of jobs, but then a lot of other jobs will be created.