A deepfake video of Warren Buffett opened the question-and-answer session at the 2026 Berkshire Hathaway annual shareholders meeting [1].
The incident serves as a high-profile demonstration of how artificial intelligence can convincingly mimic public figures. By using a synthetic version of the company's most famous leader, the meeting organizers highlighted the growing difficulty of distinguishing real human communication from AI-generated content [1], [2].
The event took place in Omaha, Nebraska, marking the first Q&A session without the presence of Warren Buffett [1]. The session was led by new CEO Greg Abel [1], [2].
The presentation began with a video of Buffett that appeared to be a live attendee. The synthetic persona introduced himself by saying, "Hi, my name is Warren, from Omaha," [1].
This unexpected opening was designed as a lesson on the risks associated with AI-generated deepfakes [1], [2]. The demonstration aimed to alert shareholders to the potential for deception, and the security vulnerabilities that arise when digital identities can be replicated with high precision [1].
As the company transitions its leadership to Abel, the use of this technology at the annual meeting underscores a shift in how Berkshire Hathaway addresses emerging technological threats. The company used the platform to emphasize that vigilance is required in an era where audio and visual evidence can be fabricated [2].
“"Hi, my name is Warren, from Omaha,"”
The use of a deepfake at a corporate governance event signals a transition in how major firms view AI risk. By simulating a breach of trust using the company's own icon, Berkshire Hathaway is moving from theoretical warnings about synthetic media to practical demonstrations of its danger, suggesting that corporate security and investor relations must now account for sophisticated identity fraud.
