Outwitting The Algorithmic Tide: Governance Risks And Director Responsibilities For AI Slop (Video)

A phrase that seems to be popping up everywhere in 2025, and which literally did not exist a couple of years ago, is "AI slop." More than a technical nuisance, AI slop (defined as low-effort, poor-quality, or factually inaccurate content mass-produced by generative artificial intelligence models) is a problem for nearly everyone, including in corporate governance. The rapid adoption of generative AI presents Canadian corporate boards with a new challenge: managing the risks associated with AI slop. It can expose organizations to heightened legal scrutiny, reputational damage, and a potential breach of a director's core statutory duties.

As AI becomes increasingly material to business operations, a board's failure to address its attendant risks, including the proliferation of slop, will make it increasingly difficult to demonstrate the requisite care and diligence of a "reasonably prudent person" under the Canada Business Corporations Act (CBCA) and its provincial equivalents.

I. What is AI Slop?

AI slop is a pejorative but increasingly recognized term encompassing the low-value, often misleading, or poorly integrated outputs of generative AI. You have no doubt run across it in your social media or even news feeds. Common features include:

• Low quality and repetitive: Content is formulaic, verbose, or repetitive, lacking authentic human insight or value.

• Inaccuracies (hallucinations): Generative AI content often creates output that presents false information or "hallucinations" as fact in a confident, authoritative tone, creating significant liability for accuracy claims.

• Bias and discrimination: AI-generated decisions or content may perpetuate biases inherited from flawed training data, leading to discriminatory outcomes.

• Unwanted or useless features: Everyone seems to add "Now with AI!" to their product descriptions these days, suggesting that the capabilities are revolutionary or necessary for efficiency. The deployment of AI, however, often includes features that do not solve a genuine business problem, leading to wasted resources and user annoyance.

The pervasive nature of AI slop means its risks are not confined to the IT department. They can touch every facet of corporate liability, from public communications to intellectual property and regulatory compliance.

II. Directors' Evolving Responsibilities in the Age of AI

Canadian corporate law imposes two overarching duties on directors: the Fiduciary Duty and the Duty of Care. The rise of AI, and specifically the risk of AI slop, places significant pressure on both.

1. The Duty of Care: The 'Reasonably Prudent Person' Standard

The Duty of Care requires directors to exercise the care, diligence, and skill that a reasonably prudent person would exercise in comparable circumstances. This standard is not static; it evolves with the times. For modern Canadian boards, this means a duty to be informed about material risks, and that now unequivocally includes AI.

• Failure to be informed: A board which remains ignorant of how generative AI is being used across the organization, or the risks posed by its poor-quality output, will struggle to defend its position under the "Business Judgment Rule" if a decision results in harm. The Business Judgment Rule is shorthand for the deference courts often grant to boards for their operational and strategic decisions. In brief, if a board makes a reasonable decision, in good faith, based on information they should have access to, a court will not normally substitute its decision over the board's (there is more to the Business Judgment Rule but these are basic factors.) At a minimum, directors must ask:

• What policies govern employee use of external generative AI tools like ChatGPT?

• How is the accuracy and quality of AI-generated public-facing content (marketing, investor communications, legal documents) being verified?

• What due diligence was conducted on third-party AI vendors to assess the risk of their systems producing AI slop?

• Risk management oversight: The board is ultimately responsible for overseeing the company's risk management strategy. This must now include a systematic framework to identify, assess, and mitigate AI-specific risks. AI slop is a direct result of inadequate technical governance, such as poor training data or deficient quality assurance models. A board's failure to mandate a robust AI risk management framework directly links to a potential breach of the Duty of Care.

2. The Fiduciary Duty: Acting in the Best Interests of the Corporation

The Fiduciary Duty requires a director to act honestly and in good faith with a view to the best interests of the corporation. The reputational and financial harms caused by AI slop can strike directly at the heart of this duty:

• Reputational damage and loss of trust: If an organization's communications or products are flooded with low-quality, inaccurate, or biased AI slop, it erodes customer, investor, and public trust. This can directly impair the long-term success and best interests of the corporation. The board's oversight must ensure management's pursuit of AI-driven efficiencies does not inadvertently destroy brand value and credibility.

• Regulatory non-compliance (AI Washing): A board has a duty to ensure the corporation complies with all applicable laws. While the Canadian government's proposed Artificial Intelligence and Data Act may be dead, it is almost a certainty we will eventually experience a future of heightened accountability and more regulation. Furthermore, the risk of "AI Washing" (misrepresenting a company's AI capabilities or the quality of its AI-generated output) is an emerging area of legal and regulatory liability, particularly for public companies. Directors must ensure that public-facing communications regarding AI are accurate and verifiable.

III. Key Corporate Risks Amplified by AI Slop

The failure to govern AI slop proactively transforms what seems like a low-level quality control issue into a significant corporate risk portfolio:

1. Intellectual property and copyright risk

AI models are often trained on vast, uncompensated datasets, raising significant copyright concerns. AI-generated content used by the corporation could infringe on third-party IP rights if the generative model was trained on protected works without a clear license. A director's failure to oversee a robust IP clearance process for AI inputs and outputs is a clear governance lapse. The current Canadian government's ongoing consultation on updating the Copyright Act to address generative AI underscores the immediacy of this risk.

2. Legal liability for accuracy, bias, and discrimination

AI slop which results in an inaccurate recommendation, such as in a product, medical, or financial context, or content that results in biased outcomes (e.g., in hiring, lending, or promotional targeting), can create liabilities under consumer protection, human rights, and tort law. The board must verify that AI systems used for material decisions are regularly tested for fairness, reliability, and security.

3. Cybersecurity and confidentiality

Unauthorized employee use of external generative AI tools often involves submitting proprietary, confidential, or sensitive customer and business data into the public model, creating a massive data breach or loss of trade secrets. Directors must ensure clear policies are in place prohibiting the submission of confidential information into externally provided AI services.

IV. Practical Governance Principles for Minimizing AI Slop

To discharge their duties and mitigate the liabilities posed by AI slop, Canadian directors should adopt a proactive, principle-based governance framework.

1. Build board literacy and strategic oversight

Directors cannot govern what they do not understand. Board education should be conducted and documented to support the defense of informed decision-making:

• Mandate AI education: Require regular, documented training on AI fundamentals, generative AI capabilities, and associated risks like AI slop.

• Treat AI as a strategic risk: Do not delegate AI oversight entirely to the IT department. Ensure it is a regular agenda item at the full board or a designated committee, such as Risk or Governance committees.

• Ask critical questions: The board does not need to be technical experts, but they must ask penetrating questions of management, such as: "What is the acceptable threshold for AI inaccuracy in our public communications?" and "How do we audit the lineage and quality of our AI training data?"

2. Implement a robust risk & quality framework

The board must require management to establish systematic guardrails to prevent, detect, and remedy AI slop. This might include the following:

3. Strengthen accountability and documentation

Clear reporting structures and audit trails are essential to demonstrate the "reasonable prudence" of the board's oversight. Steps to include might be:

• Define accountability: Clearly delineate who on the management team (e.g., the Chief Risk Officer, General Counsel, or Chief Technology Officer) is responsible for AI strategy, implementation, and risk management.

• Require regular reporting: Mandate regular, detailed reporting from management on AI strategy progress, risk assessments, incident reports related to AI errors or bias, and compliance with internal AI policies.

• Document oversight: Ensure board meeting minutes clearly reflect the nature of the AI risks discussed, the questions asked by directors, and the board's direction to management. This documentation is crucial for securing the protection of the Business Judgment Rule.

V. Conclusion

The age of Artificial Intelligence demands a corresponding evolution in corporate governance. For Canadian directors, the threat of AI slop is a tangible indication of a broader, systemic risk. By proactively addressing the governance of AI quality through continuous education, mandated risk frameworks, and rigorous accountability, boards can not only shield themselves and the corporation from increasing liability but also position their organizations to leverage the transformative power of AI responsibly and effectively. Reach out to Procido's Governance group for guidance in managing this growing part of the business landscape. 

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.