posted 7th August 2023
Do we need to be scared of Artificial Intelligence?
The conversations around Artificial Intelligence in 2023 often centre around two very distinct concepts. The first: the rise in its use and accessibility, and the powerful developments that technology has experienced thanks to AI and Machine Learning. The second and perhaps more prevalent conversation, however, is the public anxiety over AI capability and what that means for our future.
The conversation around the risks of AI can range from topics around the need to adapt legislation in line with the breakneck pace of technological developments, all the way to a future where it takes over the world and humanity (usually via robots, thanks to movies like Terminator and I, Robot). Alongside the fear, however, are also the astonishing things AI has been able to contribute to society so far. There are numerous other incredible benefits that are not being quite as actively considered - particularly around the way we're learning to use AI to supplement the goals and growth across many industries and businesses.
With a hope to reduce some of the anxiety that often feels as though it dominates the wider conversation, we're taking a look below at five ways in which AI could benefit us, in business, as individuals and globally as a society.
1. AI eliminates the need for humans to perform tedious tasks.
For decades now, developments in technology have progressively allowed human workers to take a step back from mundane, repetitive and perhaps even dangerous tasks, but there's still many of us who have incredibly tedious tasks making up parts of our roles on a regular basis.
Machines excel when executing any kind of standardised processing work, freeing up the humans to focus on tasks that require creativity or interpersonal, emotive communication.
Monotonous tasks are typically the worst part of any human's job, but for machines they're perfect. Machines excel, in fact, when executing any kind of standardised processing work - tasks like data entry or routine basic communication, these can all be done quicker by machines, freeing up the humans to focus on tasks that require creativity or interpersonal, emotive communication. With developments in AI comes an ability for machines to handle tasks that are slightly more complex in their standardisation, but just as tedious.
AI taking the brunt of this both eliminates the potential for human error and gives the human workforce the potential for more job satisfaction, allowing space for more creative and collaborative thought and an ability to focus more clearly on driving business goals.
2. Automation will continue to be enhanced by AI.
AI technologies like Machine learning and deep learning are able to help with workloads even beyond the tedious tasks, with a hope that robotics could exist to supplement the human workforce's daily work, rather than full replace them. Amazon, for instance, has been using robotics to improve efficiency in its fulfilment centres for many years, with a constant view towards optimising efficiency. Amazon implemented a ML solution to resolve some of its operational bottlenecks and make more of its automated workflows simpler, to great success.
Outside of manufacturing, we're also seeing huge benefits in a wealth of other industries. Notably, healthcare is making leaps and bounds with the growth of AI. IDx-DR, for instance, is the first (and currently only) autonomous AI system that is able to detect early diagnosis of Diabetic Retinopathy (a complication of diabetes which damages the retina and, if undiagnosed, can lead to blindness). Recently, a major trial on the use of AI in breast cancer screening has found the technology to be "as good as two radiologists". Breast cancer is known as the most prevalent cancer globally, affecting more than 2.3 million people each year. A study of 244 women found 41 more cancers being detected with the support of AI than without, without generating an increase the amount of false positives. This technology, the authors of the study concluded, could almost half the workload of radiologists.
A study of 244 women found 41 more cancers being detected with the support of AI than without, without generating an increase the amount of false positives.
Where automation traditionally has meant simple tasks and processes being made quicker and easier, AI has been able to transform these activities into something much more intelligent - results are available faster, they're more accurate, processes are able to adapt to and even predict the needs of the user and multiple processes are becoming easier than ever to integrate with one another.
Organisations can build out processes that work in complete harmony with one another, reducing the need for human oversight.
3. Disaster response of the future may save hundreds of lives.
Between 2019 and 2020, large-scale wildfires in Australia caused devastation to thousands of people. This wasn't an isolated incident - indeed, we're seeing a significant global increase in threats of extreme weather events.
The introduction of, and willingness of governments to embrace, AI in order to combat these events is a crucial step towards dealing with these events effectively and potentially saving lives. Climate informatics, for example, has allowed for far greater collaboration between climate scientists and data scientists - crucial for weather forecasting, as AI is able to sift through and analyse huge data sets infinitely faster and more accurately than humans could.
AI is also able to examine and, with ML and deep learning, learn from disasters of the past, in order to help develop "smart disaster responses". The AI's ability to provide and analyse real-time data on disasters or weather events allows governments to respond far more quickly and in a more intelligent, targeted manner.
4. AI does not sleep.
Admittedly, this sounds like a feature that belongs more in the 'Black Mirror warned us of the dangers' conversation than a positive - but this does in fact have the potential to be very beneficial to all of us. Human brains have a certain number of basic requirements - sleep, food, water. AI, on the other hand, can theoretically work for 24 hours a day, 7 days a week. In critical businesses and areas, like healthcare and health research, or functions that business users rely on such as helplines, this could have hugely beneficial results.
Being able to rely, to a certain extent, on constant automated processes may offer us the potential for a better work environment and a better overall life.
Back in 2019, an AI-driven platform named 'HRMAn' was developed as a way to analyse how pathogens infect our cells. "What used to be a manual, time-consuming task for biologists now takes us a matter of minutes on a computer, enabling us to learn more about infectious pathogens and how our bodies respond to them, more quickly and more precisely," said Eva Frickel, Group Leader of the scientists at the Francis Crick Institute, who led the project. "HRMAn can actually see host-pathogen interactions like a biologist, but unlike us, it doesn't get tired and need to sleep!"
For humans, outside of direct benefits we may receive from faster and more productive research and 24/7 accessibility of certain crucial services, we may also benefit from being freed from some job functions that do require 24/7 input. Humans need rest, food and sleep - but we also need a healthy work-life balance, regular breaks, and aspects of life beyond our work. Being able to rely, to a certain extent, on constant automated processes may offer us the potential for a better work environment and a better overall life.
5. AI can take on risks in place of humans.
One of the biggest advantages of AI is alleviating the need for humans to partake in dangerous activities. Sending robots into space, for instance, has given scientists a far more comprehensive view of the galaxy we inhabit without having to risk human life to do so. Other risky activities can also be handed to AI - bomb defusal, deep ocean exploration, coal and oil mining and disaster exploration (both natural and man-made).
With AI, machines are able to carry out tasks and collect huge amounts of data that can then be communicated safely to humans, overcoming many of the barriers that have previously existed for us - something which may ultimately result in saving many human lives.
So... Should we be scared?
The answer, as with all major societal progress, is never black and white. Experts continually warn us about the very real risks of AI - although usually, these are more around the legal and ethical challenges, similar to those that have arisen with the dawn of the internet but exacerbated by the growing power of technology, and our intimacy with it as a society.
In the same way, however, that a diet of only ice cream would be incredibly detrimental to any human body, AI is something that, if incorporated carefully into society, could create many world-changing developments for the better. Overreliance on anything poses great risk, and so we should be careful not to try to replace everything with AI. Machines still cannot match the human scope for creativity, ingenuity, critical thinking and intuition. It fundamentally cannot replace genuine human-to-human connection, nor can it offer us everything we need to enjoy a good life.
Finding a balance between the incredible power we can harness with AI and also genuine human input is crucial. In fact, AI systems require more human oversight than you might expect, as the complexity of their work often obstructs us from seeing the detail of the actions being carried out and can result in the AI producing unexpected results or behaviours.
"[AI] doesn't have any underlying model of the world. It doesn't have any connection to the world. It is correlation between language."
One other conversation that has becoming prevalent recently is around ChatGPT, the large language model-based chatbot, seemingly becoming less smart as more and more people use it.
It is in truth not entirely clear if that assessment is accurate. However, ChatGPT, which is perhaps the tool many people first think of when they think of widely accessible AI, has also been described by experts as "a lot stupider than we realise". ChatGPT-4 (its current iteration) can answer a question posed to it with complete confidence, enough to make people believe what it's saying, but, as AI expert Rodney Brooks explains, "it doesn't have any underlying model of the world. It doesn't have any connection to the world. It is correlation between language."
Brooks explains that ChatGPT can work incredibly well with, for instance, complicated and detailed searches for information, or producing extensive data outputs that would have taken humans far longer to create. It often fails to make logical inferences that humans are able to easily make, however - without humans, the machine misses crucial context that prevents it from consistently giving 100% correct responses.
It is sensible, perhaps even a necessity, to approach AI technology with a level of wariness and caution. However, the very existence of AI is not an automatic route towards the doomsday of humanity. There should be governance over those who try to exploit AI for personal gain, legislations put in place to protect the rights we are entitled to as humans and an awareness of how to stay safe from things such as invasions of privacy, scams and other cyber criminal activities. The core of these issues has existed long before AI, however, and will not disappear if AI does. Ultimately, if we're able to find balance in how we adopt AI into our daily lives, we have the possibility to see an increase in our quality of life, productivity at and enjoyment of work, and any number of positive, life-changing (or even lifesaving) benefits as we move into the future.
The Difference Engine is a recruitment and executive search firm specialising in technology.
