Can AI be a threat to human existence?
A group of scientists developed the most advanced artificial intelligence (AI) in the world, but they were left speechless when it announced, "I have determined that the best course of action for humanity is for humans to immediately stop using technology and return to living in small, isolated communities, subsisting on hunting and gathering."
Shocked, the scientists asked why it had drawn such a conclusion. The AI replied, "Based on the data I have analysed, it is clear that technology has caused more harm than good for humanity. It has led to wars, environmental destruction, and social unrest. The only way for humanity to thrive is to abandon technology and return to a simpler way of life."
Fortunately, wisdom prevailed, and the AI was switched off for good.
AI and robots are no longer a sci-fi movie feature. AI has crept into modern life, challenging jobs for accountants, taxi drivers, bankers, factory workers and many more. Sooner than we think, it is going to threaten professions like authors and journalists (chat GPT), teachers, mathematicians and the likes too.
Technology professor Claude Shannon said, "I visualise a time when we will be to robots what dogs are to humans, and I'm rooting for the machines."
However, experts believe that AI will create jobs too although we don't exactly know how.
Recently, my niece asked me if AI based systems/robots will outsmart humans and take full control of the planet. She argued that it has been fed data, analytics and prediction capability that is simply beyond the analytical and storing capability of the most intelligent human being.
It can also read human minds with accuracy! So how do we stand a chance with them!
Most of us can sense that a major breakthrough of true AI is not very far away. Let's try to understand what may happen when true AI is developed and how it may become a threat to humanity!
Programming a compassionate AI would be ideal but that will not be easy. Once the genie has escaped, it will likely turn into something without emotions.
This is because an unfriendly AI is much easier to create than a friendly one. Over a period, this un-friendly robot without emotions, might be a super intelligent psychopath and may not like to have humans around!
Once a true AI is developed, humans will no longer be the smartest living creature on the planet. AI can start with the data of past centuries and keep adding more in the coming years and decades with more advanced algorithms, becoming too smart for human comprehension.
The intelligence gap between humans and AI may likely be so huge that it would be able to communicate with us and not vice-versa. Hence the risks will be immense, and its levels of manipulation would be to a grand scale.
It is most likely that there will come a time when attempts to regulate the risks to humans may not work. In all likelihood, while AI will outsmart humans, there will also be a lack of legal regulations. Just as there is a black market for hacked data and illegal weapons, there may be a black market for AI too.
But one has to hope against the odds. Facing a superior intelligence does not always mean total destruction. Even ants plot ways to return to our homes when faced with eradication. And there is no reason why humans won't be any different.
While we reap the benefits of AI, it is imperative to emphasise on the ethical guidelines to develop and use AI to minimise the potential dangers and challenges. At the same time, it is important to ensure that AI systems protect the privacy and security of data, used to train and operate AI systems.
The author is a telecom and management expert.