AI Legal Risks: Corporate Governance & Guardrails

Corporate boards are increasingly focused on the risks associated with artificial intelligence, with a significant majority now having committees dedicated to risk oversight, according to a recent Deloitte survey.

The growing utilize of AI across industries – encompassing machine learning, Generative AI, agentic systems, and physical AI – is prompting boards to address new governance and risk management challenges. AI security breaches can compromise sensitive data, biased outputs can create compliance issues, and irresponsible deployment can have broad consequences for businesses, consumers, and society, as highlighted by Beena Ammanath, an Executive Director at Deloitte.

A February 2026 report from EY emphasizes that AI’s impact on strategy, talent, and risk necessitates a shift in board oversight. Jamie Smith and Lee Henderson, of EY, note that boards must guide companies in harnessing AI for growth, maintaining necessary skills, and ensuring accountability for AI’s uses and outputs. This includes adapting engagement with management and embedding AI into existing oversight structures.

The traditional dynamic between boards and management, often characterized by an “agency problem” – where boards rely on information provided by the management team they oversee – is being challenged by AI. According to a PwC report from November 2025, AI has the potential to address this information asymmetry, offering boards access to a more complete picture for decision-making. Paul DeNicola, Barbara Berlin, and Ariel Smilowitz of PwC state that AI can enhance a board’s ability to exercise independent judgment.

Building AI literacy among board members is considered crucial. This can be achieved through expert speakers, independent learning, and utilizing AI tools like large language models to explain complex AI concepts. Equally vital is promoting AI fluency within the C-suite, enabling executives to effectively manage the technology’s implications.

Recent events underscore the need for proactive board oversight. Instances of companies facing scandals due to inaccurate AI-generated information, alongside concerns about potential job displacement due to AI automation, demonstrate the multifaceted challenges posed by the technology. The Harvard Law Review has also noted the potential for “amoral drift” in AI corporate governance, particularly within AI startups reliant on scarce talent and resources.

The EY report highlights a shift towards transformative AI applications, moving beyond simple efficiency gains to focus on process reinvention and new business models. This requires boards to challenge conventional assumptions about their engagement with management and the broader world.