Unlock The Full Potential Of Artificial Intelligence

The past year has been a learning experience for all of us, and collectively we’ve managed incredible things. Out of necessity, organizations have massively increased their speed of digitization and uptake of game-changing technologies such as artificial intelligence (AI) and cloud computing.

As 2020 draws to a close, it isn't uncommon to hear it said that businesses had made several years’ worth of progress in their journey of digital transformation in just the last 12 months. Clearly, there’s nothing like danger for focusing the mind and encouraging human beings to come up with innovative solutions.

This increased uptake is leading to a lot of successes – particularly at the pilot stage. But inevitably, when it comes to deploying them at a mass scale in the real world, not all of them get the chance to make a real mark.

Projects fail to make it past pilot when it becomes difficult to demonstrate their real-world value, or fizzle out when our enthusiasm is directed towards something different. One key thing I have learned is that the maturity of understanding within organizations, when it comes to hugely disruptive technologies like AI and cloud computing, is still catching up with the appetite and enthusiasm. It’s simple enough to set up a pilot, but scaling to a fully deployed application is a different story.

So, once you know that AI has the potential to solve a problem for you, how do you go about making sure your projects or applications don’t languish in the pits of pilot hell? You may be able to use AI to prove a concept, but without that organization-wide maturity, it may be difficult to scale and deploy operationally in a way that will ensure it really makes a difference.

Clues can always be found by looking at instances where it was done successfully. At University College, Dublin, researchers working with IBM, combined digital surgery with artificial intelligence, using videos of surgical procedures. This has led to the creation of what can be thought of as a “digital biopsy” tool that can be used to identify cancerous tissue from camera images taken inside the human bowel. One of the challenges here was that although traditional image classification algorithms are now very proficient when it comes to identifying clearly distinct shapes, for example, cats depicted in digital images, but more nebulous shapes – such as cancerous tissue clusters amidst our innards – have generally proven more elusive.

 Advances pioneered by this team mean that they have been able to create algorithms that have a better than ever before chance of spotting and correctly diagnosing these tumors. And now they've proven successful at identifying cancerous bowel tissue; the technology is ready to be scaled out and put to work helping surgeons in the real world.

The next step for this project will be to train it to work on 3D models rather than simply flat video images. This will help us understand more about the way tumors act as they move through the body, and lead to better rates of detection and ultimately, patient recovery.

Another potentially hugely transformational AI use case is seen in the Mayflower autonomous ship project. This crewless, self-piloting research vessel was launched from Plymouth, England this year to mark the 400th anniversary of the original Mayflower voyage. However, rather than simply being a commemoration of the past, the launch of the ship marks a step into the future, with scientific research considered to be one area where autonomous, AI-piloted vehicles could be hugely beneficial. 

Much of the space of traditional vessels is taken up by amenities necessary for the comfort (and sometimes survival) of human crews. Removing cabins, washrooms, food preparation and storage areas and lifeboats means autonomous research vessels can be far smaller and lighter than they would need to be if humans were on board. The need to support and provision humans also makes scientific research voyages very expensive. Removing these considerations will lead to more cost-efficient scientific exploration and discovery. Currently, the ship is tooled to take part in studies involving the monitoring of microplastic pollution, sea water quality and marine mammal populations. But it can quickly and easily be repurposed to any kind of oceanographic research, meaning it could potentially revolutionize the field.

Both of these projects have the distinction of making it beyond the pilot stage and to the point where they could be used to make a real impact. And this is really due to several other characteristics that they both share.

Both projects were conceptualized to help with a real challenge that can only be solved by more intelligent analysis and understanding of data – whether it’s medical imagery from internal camera examinations, or video from shipboard cameras used to plot a safe passage for the autonomous vessel. Neither was a case of deploying AI simply for the sake of being able to say “look how clever we are, we’re using AI!” Rather, AI offered the only viable solution to a real operational problem, so was deployed out of necessity.

Another factor these projects had in common was buy-in from the people whose jobs will be transformed by the arrival of this technology. Surgeons understand the need for a tool that can assist them with carrying out the critical job of reviewing medical imagery during operational procedures, where they may already be using every ounce of concentration to ensure the best patient outcome. And although many marine biologists love the ocean and love being at sea, they understand the efficiency benefits that will come with the ability to deploy robotic research vessels, such as surveys that can cover far greater areas in far less time than is possible using manned voyages. 

Trust is an essential element of securing this buy-in – it’s one thing to be able to prove that your project can drive real-world impact if deployed at scale, but it’s also essential that you can show it will do so in an ethical and reliable way. For this reason, putting measures in place to mitigate against bias is critical, as is transparency over how your data is collected and processed.

I believe it’s been firmly proven that no matter what your job is, in any industry, AI has the capacity to revolutionize the way you work and the way your organization does business. As use cases pile up, however, it’s becoming clear that it takes more than a good idea and cutting-edge technology to realize the full effects of transformation. It’s no longer the technology that’s holding us back - the practical barriers that need to be overcome between pilot projects and game-changing real-world deployment are largely organizational or human-shaped. But by paying attention to those who have succeeded, and putting plans for overcoming those barriers at the heart of your strategy, you’ll give yourself the best shot possible at making a real difference.

If you would like to learn more about any of the examples or topics mentioned here, then join me at IBM’s Cloud and AI Forum – Unlock the full potential of Cloud and AI. You can join us live on Nov 26th or, once registered, can watch any of the sessions in your own time.

Thank you for reading my post. Here at LinkedIn and at Forbes I regularly write about management and technology trends. I have also written a new book about AI, click here for more information. To read my future posts simply join my network here or click 'Follow'. Also feel free to connect with me via Twitter, Facebook, Instagram, Slideshare or YouTube.

About Bernard Marr

Bernard Marr is an internationally best-selling author, popular keynote speaker, futurist, and a strategic business & technology advisor to governments and companies. He helps organisations improve their business performance, use data more intelligently, and understand the implications of new technologies such as artificial intelligence, big data, blockchains, and the Internet of Things.

LinkedIn has ranked Bernard as one of the world’s top 5 business influencers. He is a frequent contributor to the World Economic Forum and writes a regular column for Forbes. Every day Bernard actively engages his 1.5 million social media followers and shares content that reaches millions of readers.