Harnessing AI for social good

It seems like everyone is talking about AI and offering up solutions that could potentially change the world. Unfortunately, there are still lots of myths circling around AI, about what it is and isn’t. AI, for example, is far more than advanced, sophisticated analytics. This has led to confusion among enterprises, business leaders and governments about what AI might actually achieve for the digital economy and wider society.

I was fortunate to work on a global AI project for an international software company some years ago and have seen first-hand the potential the technology has for social good. But, I also understand that we need to put its capabilities in perspective. It isn’t a silver bullet to the world’s woes.

Uncovering AI

We have to remember that AI is a computer engineering discipline. As Alexander Linden, Research Vice President at Gartner, recently pointed out, some forms of machine learning (ML), a sector of AI, “may have been inspired by the human brain, but they are not an equivalent.” Yes, AI can mimic human behavior. But, AI has a very long way to go before it can use content memory and think like us.

AI can reduce bias, but it can’t eliminate it entirely as it can creep into training data algorithms. Research at MIT found that ML algorithms can actually discriminate on race and gender, for example, if it is fed biased data.

AI has the ability to do a lot of good, but biased AI systems could cause problems in the future, fueling and aggravating existing prejudices and social inequalities if not properly controlled. In this area, we will undoubtedly see growing government legislation around algorithms that are set to be a complex legal battle.

AI for social good

The good news is that beyond big business, organizations are already using AI to address pressing social issues. Technology heavyweights such as Microsoft, Facebook, Intel and Google have been quick to jump on board and already have exciting projects up and running that are already making a difference to people’s lives.

India, for example, has a dire shortage of safe blood for transfusions. Facebook developed an AI tool that enables individuals and organizations to link up in giving and receiving blood, and the tool is now available in Pakistan, Bangladesh, and Brazil as well as India. AI recognizes when the content of a Facebook post is related to donating blood and automatically sends the person or organization a message and invites them to participate. Currently, 35 million people have signed up to be blood donors, and countless lives have been saved.

Microsoft’s AI for Earth is another innovative program designed to put AI tools in the hands of those working on global environmental issues. AI partners include OceanMinds, which is using AI to protect biodiversity and increase the sustainability of global fishing, while WildMe is using AI to identify animal species on the verge of extinction. These are just a few of the many inspirational AI projects underway that will change our planet for the better.

In the significant challenges faced by companies this year, it is increasingly critical to innovate with tools and processes harnessing AI and offer overall better digital accessibility. At Orange, we recognize the benefits of teaching AI to optimize social good results and avoid biases and inequality. You will find useful articles about AI and related innovation research on our Orange website.

Collaborating for common good

Real-life examples of AI are already all around us, from diagnosing diabetes and certain cancers to predicting the risk of natural disasters such as earthquakes and Tsunamis. We are seeing AI track everything from diseases to levels of pain.

Orange is developing and collaborating on AI solutions for critical infrastructures and environmental issues. Our innovative solutions, such as Orange Flux Vision formulate accurate statistics on mobility patterns in real-time. Flux Vision is being used by the World Health Organization for an epidemiology project in the Congo in a bid to control diseases and other factors relating to health. The technology was successfully used in a similar project by Dalberg Data Insights, which links data solutions to international development challenges in the Republic of Guinea.

At the Centre Hospitalier Universitaire du Nice in the South of France, Orange has been working on creating a solution for pain assessment using facial expressions and AI technology. This is proving invaluable in pediatrics, for example, particularly post-surgery, where babies and small infants can’t communicate how they are feeling.

For AI to deliver further on its potential to change our world for the better, its use for social goodwill needs to be scaled up - especially around the areas of data accessibility and skillsets. AI talent is still in very short supply, which is ultimately slowing down breakthroughs in the field of medicine for example.

Multi-sector collaboration is to my mind essential for the safe and ethical advancement of AI for social good. Stakeholders in both the public and private sectors will need to collaborate together more to ensure that AI achieves attainable goals for social good. This means sharing data and skills.

AI can have a very positive impact on our world, changing the lives of millions for the better. It is essential, however, that we have a framework for responsible innovation moving forward if we are to harness the power of AI for social good and not end up with more problems than solutions.

Orange Business is committed to serving our customers on their journeys. You can read more about it from one of our latest press releases. In the significant challenges faced by companies this year, it is increasingly critical to innovate with tools and processes harnessing AI and offer overall better digital accessibility.