In June 2026, Bangladesh Bank took the unusual step of prohibiting its staff from entering sensitive financial information, policy drafts, or internal notes into generative artificial intelligence platforms such as ChatGPT, Gemini, and DeepSeek. This directive came after the regulator identified instances where employees had processed unverified financial data and confidential files through external AI systems, raising concerns over potential cross-border data leaks.
This move reflects a broader, global challenge facing corporate and financial institutions as they rapidly adopt AI technologies without fully understanding the associated risks. Across major financial centers worldwide, including Dhaka’s Gulshan district and Wall Street, executives are under intense pressure to demonstrate digital innovation. However, experts warn that this rush can lead to significant governance failures when human oversight does not keep pace with technological adoption.
Recent data underscores a concerning gap between trust in AI outputs and the preparedness of organizations to validate such information. Approximately 84 percent of senior executives reportedly place confidence in AI-generated results without sufficient review, while only 11 percent of companies have the necessary data quality to support these decisions. This disconnect has contributed to significant liabilities, with about a quarter of business leaders acknowledging that critical AI errors—often referred to as “hallucinations”—have slipped past internal audits, sometimes reaching boards or public financial disclosures before correction.
Consequences of these oversights are becoming more pronounced. AI-related securities lawsuits have increased internationally, accounting for nearly three-quarters of all disclosed investor losses linked to technology errors. Some analysts describe these incidents less as technological malfunctions and more as failures in organizational culture, human judgment, and talent management.
Emerging trends such as “Agentic Finance”—where autonomous AI agents manage capital flows in real time—amplify these risks. Industry observers have warned that such systems could potentially trigger rapid “AI bank runs,” causing liquidity crises that might drain trillions of dollars from conventional banking within minutes. These dangers intensify if human operators lack the critical thinking skills necessary to intervene effectively when anomalies occur.
Despite the mounting risks, the financial return on AI investments remains underwhelming. Studies indicate that around 95 percent of enterprise AI pilots fail to scale or produce tangible financial benefits. Experts attribute these shortcomings to leadership approaches that treat AI primarily as a marketing tool rather than integrating it thoughtfully into existing infrastructure while developing requisite human capabilities.
Human resources professionals are called upon to play a vital role in addressing these challenges. Experts advise redefining hiring criteria to prioritize cognitive skepticism and risk management competencies over mere technological buzzwords. They also recommend implementing stringent “human-in-the-loop” protocols to ensure that no AI-generated financial or strategic outputs reach decision-makers without rigorous human validation. Additionally, corporate cultures should be reshaped to reward accuracy and operational resilience rather than the adoption of flashy, unproven technologies.
In summary, while AI offers potential efficiency gains, it cannot substitute human judgment, ethical oversight, and accountability. As organizations integrate advanced technologies, there is an urgent need for strategic caution, enhanced governance, and a renewed focus on human expertise at the highest levels of management.