Can AI be Smarter Than The Smartest Human?

 Elon Musk has made a bold assertion about AI research, saying that Artificial General Intelligence (AGI) will outperform human intelligence by next year or 2026. In a recent interview on X, Musk emphasized the need to have enough electricity to facilitate AI growth. He also discussed the difficulties experienced by his firm, xAI, notably in training the next iteration of their AI chatbot, Grok, due to a lack of powerful chips.

This prediction comes amid Musk's continuing legal struggle with OpenAI, a company he co-founded, over claimed deviations from the company's initial mission of employing AI for the greater benefit. Musk's comments reflect the growing debate over the ethical aspects of AI research and its potential societal consequences. However, he has previously commented on AI and humans. 

In reaction to a recent podcast presented by Joe Rogan, Musk wrote on X that by 2029, AI will most likely be smarter than all humans combined. During the podcast, the guest, Ray Kurzweil, an American computer scientist, stated that while people may expect AI to take another 100 years to become smarter than humans, he believes AI could reach this milestone even sooner, possibly within the next five years -- a prediction similar to Elon Musk's.

However, Musk feels that AI could be detrimental to humans. However, he believes the benefits of AI outweigh the risks, according to Business Insider. The wealthy entrepreneur did not go into detail about how he arrived at the risk estimations.

How Intelligent Are AIs?

Artificial intelligence has advanced significantly since its release. Many AI specialists believe that human-level machine intelligence (HLMI) will arise within 45 years, with a 50% probability, and within 9 years, with a 10% chance, indicating that they have a chance to learn how to duplicate the human brain. Interestingly, AI systems have already outperformed humans in several categories, for example:

Activities include strategic games, surgeries, and driving cars.

However, AI cannot still think independently or understand human behaviour and its complexities, much alone making decisions based on common sense. AI is programmed by humans, and they currently lack emotional intelligence and common sense. Artificial intelligence (AI) is simply well-trained to perform certain jobs under defined constraints, rendering all popular beliefs irrelevant.

Can AI Replace Humans?

While it is never truly black and white, AI is unlikely to completely replace humans. Instead, most experts believe that AI will augment rather than replace human capabilities.

And, as previously discussed, AI is still a long way from replacing people or their intelligence. While AI can do jobs fast and effectively, it lacks the depth of general intelligence, creativity, and social understanding that humans have.

However, it is projected that Artificial Intelligence (AI) will eventually evolve into artificial general intelligence (AGI), acquiring cognitive traits similar to those of humans, such as creativity and emotional awareness. 

Can Artificial Intelligence Take Over Human Jobs?

While AI lacks emotional intelligence and the common sense of people, it is ultimately a machine. And yeah, it is quite fast. So, in terms of all automated chores, calculations, repetitive operations, and so on, AI can and will most likely replace humans, including driving. So, what jobs are left out?

Jobs requiring complex decision-making, strategic thinking, and interpersonal skills are unlikely to be displaced by AI.  It still struggles with activities that require human-like cognitive ability, creativity, and emotional intelligence, consequently altering the professions that humans can perform.

How Long Will It Take for Artificial Intelligence to Become Smarter Than Humans?

Predicting how long it will take for AI to outperform humans entails tremendous uncertainty and differs widely across specialists in the subject. The timescale for reaching Artificial General Intelligence (AGI), in which AI matches or outperforms human intelligence across a wide variety of tasks, is extremely uncertain. These predictions are influenced by technology developments, finance, ethical considerations, and societal consequences. Here's a summary of the many perspectives:

Optimistic Estimates

Some technologists and futurists believe that AGI will be developed within the next few decades. For example, Ray Kurzweil, a well-known futurist and Google Director of Engineering has proposed that AGI might be realized by 2029, with the possibility for AI to outperform human intelligence shortly thereafter. Such optimistic estimates frequently rely on the quick speed of current advances in machine learning and processing capacity.

Pessimistic or Cautious Estimates

Other academics are more cautious, predicting that AGI may not be realized for decades, if at all. This viewpoint is based on the enormous complexity of human intellect, as well as the major technical and ethical issues that have yet to be addressed. 

Surveys of AI Researchers

Surveys of AI researchers reveal a diverse range of forecasts. AI Impacts conducted a study in 2016 and produced a median prediction of 2040 to 2050 for AGI, with significant variation among respondents. Similarly, a survey presented at the 2016 Puerto Rico AI conference revealed a 50% possibility of AGI by 2050. However, the same surveys also demonstrate that projections vary greatly, highlighting the field's high level of uncertainty.

Conclusion

Unexpected advances in AI research or computational technology (such as quantum computing) may have a substantial impact on the timetable. Similarly, legislative acts, ethical reasons, or serious societal problems may impede the development of AGI.

While there is no consensus on when AI will surpass human intelligence, a variety of expert projections indicate that it will happen within this century. However, this is theoretical, and the real timescale will be determined by a variety of factors such as technical advancements, societal attitudes, and regulatory frameworks. The creation of AI smarter than humans not only presents a technical challenge but also raises fundamental ethical and societal concerns that humanity must carefully address.