Who’s Accountable When AI Fails?

Business leaders need to assess accountability at every level to effectively manage AI risks, writes Wharton’s Cornelia Walther.

KEY TAKEAWAYS

• When AI systems fail, responsibility quickly cascades across multiple levels — from developers to corporate boards to regulators.

• Effective AI accountability requires a four-level approach that integrates individual mindsets with organizational governance.

• Organizations that implement proactive accountability frameworks face fewer legal challenges and build greater public trust, which determines whether their AI systems scale or stall.