The deep dive: Adopting AI responsibly

By Puja Sharma

The deep dive’ is our bi-weekly exploration of a relevant topic, hot trend, or new product. For Prime subscribers only.

How does it work?

Fueled by the pandemic, the adoption of artificial intelligence (AI) technology is on the rise, with more than 72% of global enterprises have embarked on their AI journey, according to Everest Group. Global spending on AI services is expected to accelerate 32%, from $25b in 2019 to $95b by 2024.

While the transition from AI Proof of Concepts (POCs) to production was already underway, the pandemic further accelerated this shift. Before the pandemic, enterprise application of AI was focused on improving the efficiency and effectiveness of enterprise operations; however, during the pandemic, there has been a visible shift towards leveraging AI to improve stakeholder experience. A new set of use cases, such as touchless AI operations, social distancing solutions, and behavior drift analysis have further increased enterprise spending on AI.

Who is under the radar?

As enterprises aim to scale the technology, they are presented with numerous challenges. A burgeoning skills gap, the lack of proven return on investment, concerns around privacy and regulations, and the black-box nature of the technology act as journey impediments. Among all, talent continues to remain the key challenge for firms.

Banking, Financial Services and Insurance (BFSI), and healthcare and life sciences lead AI adoption, followed by manufacturing. North American enterprises contribute nearly 51% of the revenue for AI services, followed by European firms. Asia Pacific (APAC) enterprise spending on AI has been increasing in recent years, led largely by Chinese and Japanese firms, owing to a significant government push for adoption. Enterprise AI spend is highest amongst customer service as well as sales and marketing front-office functions, followed by human resources and finance and accounting in the back office.

Privacy and security concerns were cited as the greatest barrier to adopting and incorporating AI technologies by bankers surveyed. Especially relevant to smaller organizations whose parent company has assets below $10bn. As a result, larger banks are more likely to struggle with regulatory compliance, complexity, or technological limitations. Banking provides fertile ground for artificial intelligence (AI) thanks to its numbers-based, data-driven nature. Using AI for routine tasks in banks has initially proved to be low-risk and incrementally beneficial. AI is also offering banks transformational opportunities for product innovation and new business models.

Why does it matter now?

To overcome the expanding talent gap, the enterprises should invest in democratizing AI, ensuring that the technology is accessible to all within the organization. Investments in data and AI literacy, self-service no-code and low-code tools, and automation-enabled machine learning will be key to success. This democratisation effort also will require contextualisation, change management, and governance to ensure responsible and successful use of Artificial intelligence as access expands within the enterprise.

Security and privacy breaches pose the greatest risk in association with the adoption of AI, closely followed by the failure of AI. Bias and trust are prominent barriers and the issues of “explainability” are gaining prominence. Technology decision-makers prefer responsible for AI adoption to balance and business benefit against increasing complexity and risk. The data by IBS intelligence shows what are the greatest risk to banks associated with adopting AI in 2022.