Two AI Stumbles in One Week: Why Values-Based Leadership Must Be Distributed In the space of just three days this month, two major AI companies highlighted a critical weakness in how leadership is practiced today. On July 10, Apple filed a lawsuit alleging that OpenAI's hardware division, led by former Apple executive Tang Tan, engaged in a coordinated effort to steal trade secrets. Just three days earlier, Meta launched a feature in its new Muse Image tool that allowed users to generate AI images of people by simply tagging their public Instagram accounts — with the feature turned on by default and no notification to the person whose likeness was used. Meta removed the feature days later after widespread backlash. These incidents are very different in nature. The allegations against OpenAI, if true, point to something deeper than a miscalculation. They suggest a potential breakdown in ethical boundaries at a leadership level inside a high-growth organization. Meta's case, while serious, looks more like a failure of judgment — moving fast under competitive pressure without properly accounting for user expectations around consent and likeness rights. What connects both cases is this: when the financial and competitive stakes are extremely high, internal pressure can distort decision-making. Even experienced leaders are influenced by the culture and incentives around them. Company culture is not neutral — it either reinforces ethical guardrails or makes it easier to overlook them. In my book AI L4IR: Leadership and the 4th Industrial Revolution, I argue that ethical leadership in this era cannot remain concentrated at the top. It must be highly distributed — embedded in how product teams, engineers, recruiters, and middle managers make decisions every day. When billions of dollars and strategic positioning are on the line, relying on a small group of senior leaders to catch every ethical risk is unrealistic. Organizations that build cultures where people at all levels feel empowered (and expected) to ask "Should we?" — not just "Can we?" — will be far better positioned to innovate responsibly and maintain trust. I explore both incidents and the broader leadership implications in more detail here: https://kentkaufman.substack.com/p/two-ai-stumbles-in-one-week-why-values How are you thinking about building stronger ethical decision-making into your teams as AI capabilities continue to advance?

Human Judgment