The trend: Advances in AI are making the technology ubiquitous, especially in the financial sector, but issues regarding ethics and reliability loom large. Surveys reveal that C-level executives in multiple industries aren’t too keen to start using it themselves.
- AI is routinely used for tactical, lower-level decision-making, yet it isn’t standard practice for higher-level strategic decisions where executives prefer to trust their gut, per Harvard Business Review.
- A survey of 2,190 executives from nine countries, representing banking/financial, insurance, telecom, retail, HC/LC, and government, found that 67% of CEOs often prefer to make decisions based on their intuition and experience rather than insights from data analytics, per KPMG.
- In a Deloitte survey, 67% of executives said they weren’t comfortable accessing or using data from advanced analytic systems.
- 42% of data scientists from a cross-section of industries said their results aren’t used by business decision-makers.
AI is easy to implement, but safeguards aren’t: From assessing loan applications and determining insurance premiums to deploying customer service chatbots, AI is affecting our world, with much of it occurring in the background.
However, recommendations to surmount the issues that make executives wary—like creating more reliable AI models, avoiding data biases, ensuring ethics, and explainability—aren’t easily realized.
- The rise of AI's complexity, including Deep Generative Models (DGMs) created by the fusion of neural networks and generative modeling, is making it even more opaque, limiting developers’ ability to curb bias and ensure consistency of real-world results.
- Efforts to teach ethics to AI, effectively endowing it with a superego, result in such a quandary that it might be impossible. AI expert Lance Eliot stated, “The better the AI becomes, the worse the AI ethics problem becomes.”
- AI’s prowess is downright sinister at times, and ironically, despite its promise of improved cybersecurity, AI is making banks vulnerable to Russian hackers.
The bigger picture: The widespread concern about AI is at odds with the cavalier approach to its implementation and makes last year’s dismissal of AI ethicists jolting. AI innovations suggest the technology could be applied in potentially infinite ways, but that isn’t necessarily the best use case.
- AI’s implementation is already affecting the lives of millions, and its use for high-level strategic decisions comes with an outsized risk.
- The technology doesn’t have to be used in every conceivable way for society to reap the benefits.
- AI in conjunction with robotics and automation could potentially perform all occupations, but the reality is that many people value uniquely human contributions.