- Where to Play 2021+
Artificial Intelligence has the potential to contribute an estimated $15.7 trillion to the global economy by 2030, according to research from PwC. This exponential growth represents the most significant commercial opportunity in today’s fast-paced and competitive economy.
It’s no longer a question of whether or not organizations should apply AI solutions—which have demonstrated value in terms of speed, accuracy, and resilience. But rather, the discussion centers around how it should be applied and what responsible AI actually entails.
In our recent Innovation Rockstars podcast episode: How to Make the World a Better Place with AI, Dr. Christian Mühlroth covers this topic with Michael Berns, AI & FinTech Director at PwC.
Berns presents three innovative use cases in which AI is harnessed for good. In each, an AI model has been trained to detect a specified pattern within a huge volume of unstructured data. He likens it to finding a needle in a haystack. And in each of these cases, the needle holds the key insights to help the custodians of the AI model make informed decisions, prepare preemptive responses, and provide an enhanced level of protection.
AI and the financial sector are ideal bedfellows. AI technology such as Natural Language Processing (NLP) and predictive analytics have the capacity to transform customer engagement and workforce productivity for financial service providers. But the application of AI goes far beyond process automation, emphasizes Berns.
Banks and other financial institutions are building AI solutions that assist in regulatory reporting and compliance—sometimes referred to as RegTech, or SupTech when employed by financial supervisory agencies.
This entails the detection of anomalies in transactions or communications that point to potential threats like collusion, money laundering, front running, or other financial crimes. As this type of criminal activity grows increasingly advanced, the financial sector must keep pace and implement smarter mechanisms to accelerate risk detection and mitigation.
With many global healthcare systems under pressure—even prior to the pandemic—any solution that saves time and resources has the potential to save lives. Berns cites the development of AI models that are able to identify connections between different complex sets of data and contextual information. Healthcare providers can use these insights to help inform their diagnoses and treatment plans, e.g., in cancer cases.
Berns refers to this as a rare win-win-win situation. AI applications in healthcare enhance patient care, making it more personalized, targeted, and efficient. Hospitals can be better run and managed, with resources and capabilities optimized. Meanwhile, the insurance industry is able to make more accurate predictions to ensure the best coverage and cost for their customers.
AI will be increasingly integral in preventative healthcare and disease management. There are examples worldwide demonstrating the role of AI in helping to curb the spread of COVID-19—monitoring symptoms and cases, prioritizing patient support, and, somewhat controversially, enforcing social distancing measures.
Combatting human trafficking and exploitation has posed a challenge for governments worldwide. There is a lack of comprehensive and reliable data for anti-trafficking organizations to understand the scope of the problem, making it difficult to devise effective measures to identify and prevent trafficking.
This is where AI can assist. Similar to the software developed to help law enforcement agencies detect terrorism, these NLP models can derive intent from communication in online forums, chat groups, and classified advertising websites. It pulls out patterns and profiles that suggest a possible exploitation case, allowing law enforcement to direct their resources for further investigation.
Facial recognition technology is another AI-driven tool in the fight against human trafficking. It uses deep learning algorithms to match missing persons with online images showing victims of trafficking.
Berns has done previous work with Thorn, a tech-oriented anti-trafficking organization. Thorn uses AI solutions to help empower law enforcement, reducing the critical time spent searching for juvenile victims of trafficking by 60%—resulting in the identification of 9 children per day.
The examples above exemplify a few of the ways that AI can be used to make the world a better, safer, and more just place. However, Berns, who admits he’s been referred to as an AI evangelist, knows that the adoption of AI is not without its problems.
Berns says that while most business leaders claim they are interested in integrating AI solutions into their organizations, there are gaps in readiness. Businesses in China and the U.S. are much further ahead, in general, investing more money into developing AI tools, adapting processes, and reskilling their workforces.
The geographical disparity in terms of AI preparedness also gives rise to discussion of AI equity. Should access to advanced technology like AI remain limited in low- and middle-income countries, the digital divide will continue to widen, leaving these economies ill-equipped to compete globally, build resilience, and increase capabilities within their workforces.
Effective AI applications are highly dependent on a critical mass of data. Meanwhile, governments worldwide are progressively tightening regulations around the procurement and manipulation of data—especially where it is related to personally identifiable information, such as in healthcare. Ensuring data compliance can be a lengthy process, and collectors of this data must clearly communicate their intentions, as well as the benefits of opting-in.
As with any new, disruptive technology, there exists a tension around the adoption of AI. People fear that AI and its branches like robotics and expert systems will replace humans and create massive job loss. And while it’s inevitable that AI will lead to the migration and indeed obsolescence of certain roles, Berns asserts that it will equally create opportunities for new roles within improved organizations.
Berns would say, unequivocally, yes. He admits, though, that AI is just a tool, and is neither inherently good nor bad. Much like social media, or the internet in general, it is how this tool is used that makes all the difference.
Berns concludes, “Don’t be afraid of AI and disruptive technologies in general. They are here to stay; they’re not going anywhere.” And as AI becomes progressively ubiquitous in our lives, the mandate for responsible or ethical AI will grow in importance.
To gain strategic advantage, organizations must strive to be at the forefront of integrating bias-free algorithms and advanced machine learning into their processes. When these AI applications are configured to help deliver on an organization’s purpose—combining the best of machine intelligence and real human vision and empathy—here is where AI does good.