Whether it’s adverts for the Amazon Alexa, Sophia the robot tweeting or ‘Black Mirror’ throwing its viewers into a hellscape where personal security robots routinely hunt and kill humans, our society has Artificial Intelligence (AI) on the brain. With its applications seemingly endless, an AI takeover does seem inevitable. However, it has some drawbacks that may stop this superintelligent revolution in its tracks.
Ostensibly, AI looks overwhelmingly positive. It can adapt, act, sense, learn and reason in response to the real world, which means that AI systems can act in the same way a human can, but have the intelligence of a computer and can solve and reason far faster than any human could. This is incredibly useful, as it means that, developed fully, AIs would take over potentially dangerous tasks, thus meaning that workplace deaths and accidents would become virtually non-existent.
In addition, as AIs can ‘think’ and ‘see’ for themselves, they are invaluable in helping those who are disabled, and AI technology is already in use for those who are blind or partially sighted. ‘Seeing AI’- an app produced by Microsoft- is able to identify the user’s surroundings, read text and currency, and even detect facial expressions.
Likewise, AIs would be useful in looking after those who are not able-bodied, the sick and the elderly as, unlike a human, they cannot get sick themselves, do not need breaks, and are available 24/7. Their superior reasoning skills also mean that AIs can be applied across businesses and agriculture; for example, an AI called ‘FarmLogs’ is currently used by one out of three farmers in US, to monitor their fields and therefore maximise their crop yields and profits.
AI has the potential to benefit so many people in so many ways, and these ways will likely increase as more research and development goes into superintelligence.
AI systems can act in the same way a human can, but have the intelligence of a computer and can solve and reason far faster than any human could.
However, whilst AI is incredibly useful, it can also be dangerous. Tesla recently suffered massive setbacks when their Model X self-driving car struck a concrete median and killed the driver, Walter Huang. The crash itself has been thought to have occurred as the navigation system of the Tesla misread the lines on the road and drove the car into the median.
Tesla released a statement to ABC news in which they said that ‘the only way for this accident to have occurred is if Mr Huang was not paying attention to the road ahead, despite the car providing multiple warnings to do so’. Even if Mr Huang is at fault for failing to watch the road, it does not take way from the fact that AI itself is flawed as it failed to recognise the lines on the road– something that is arguably crucial for a self-driving AI to do.
In addition, whilst he was aware that the AI was not perfect, Mr Huang should have been able to take his hands off the steering wheel for more than six seconds without a crash occurring– which is unfortunately what happened. If we are to put our faith in AI, as those who will need self-driving cars in order to get around and will be unable to intervene should the car’s navigation system fail (for example, the visually impaired), then mistakes like these cannot afford to be made.
AI itself is flawed as it failed to recognise the lines on the road– something that is arguably crucial for a self-driving AI to do.
Overall, I believe there is a bright future ahead wherein AI is developed and applied in ways that help both businesses and people with their problems. However, I also believe that AI in its current state can be flawed when it comes to doing things ‘better than a human’, e.g. driving, proven by the Tesla accident.
Whilst AI is undoubtedly great in many ways, it does have a long way to go before it is fully integrated into society, especially if people are to be reliant on it, whether this be in businesses or in their everyday lives. Still, it is undoubtedly exciting to consider the potential benefits that we can reap from AI.
If you want to read more about AI, specifically some of the revolutionary tech involved read this article where Jaysim Hanspal looks into some of the tech that will make AI possible.