Artificial Intelligence has potential to improve or destroy our future

AI is a powerful innovation that could transform humanity. [iStockphoto]

A few days ago, The African Union Commission fell victim to cybercrime. Cybercriminals faked Secretary-General Mohammed Faki's voice and set up a series of meetings with global leaders.

It was not only embarrassing, but also caused a diplomatic scare since the pranksters held video calls with several European leaders. All this happened thanks to Artificial intelligence (AI).

AI technology has provided huge improvements to life and continues to transform humanity as it grows. Improvements have already been felt in medicine, research, agriculture, and service industries. Mustafa Suleyman, co-founder, and former head of Applied AI argues that in just five years, AI will be able to make phone calls to humans to negotiate a business proposal.

He further predicts that it will be possible for AI to contact other AI to plan a shipment, pay freight, taxes, and insurance. While the benefits, both now and in the future, remain appealing, it will be naive to ignore the potential dangers of AI.

AI has always been associated with job losses. Investors like Mustafa argue that jobs will not be lost but rather change into more sophisticated jobs, like machine learning. On the other hand, critics like historian Yuval Noah Harari, speak of a dangerous scenario where AI will create a small number of sophisticated jobs in a few places like California's Silicon Valley, but lead to losses in manufacturing jobs in developing countries like Bangladesh and Pakistan.

A bigger risk attributed to AI, is the possibility of development of self-controlled weapons, with the ability to harm and kill without human intervention. This could be the new arms race increasing the risk of nuclear war.

Another danger is that AI technologies will allow governments and business to come up with mass surveillance systems that could be used to track and monitor our every move. This would lead to a loss of freedom and privacy and other resultant ills that come with invasion of privacy such as blackmail and related crimes.

As was the case with the AU Secretary General, AI has already been used to create manipulated sounds, videos, images, and recordings to make it look like one is saying or doing something that they never in real sense said or did. This could be misused to spread misinformation, damage people's and organizations reputations, terrorize and blackmail or even interfere with political processes like elections and referendums.

In addition to these specific dangers, there is the good old concern that AI could become too intelligent to be controlled as it will have surpassed human intelligence. This might be a threat to humanity.

While this scenario has been explored in movies and science fiction stories, we need to remember that the future remains unknown and a global pandemic like Covid-19 had similarly been predicted by many a film director. Additionally, some of the world's leading AI researchers have warned that this is a real possibility that ought to be taken seriously.

It is paramount to take steps that would mitigate against these dangers. Investment in research on aligning AI with human values should be a priority. Equally important is the development of treaties that would ban or prevent development of autonomous weapons and mass surveillance. The public should additionally be made aware of the potential risks of AI so they can make well informed decisions.

While AI is a powerful innovation that could transform humanity, it is important that its dangers are not only acknowledged but deliberate efforts made to mitigate against them.