A developer named Kun Chen built AI chatbot versions of four prominent tech CEOs, dropped them into a single group chat, and instructed them to debate the future of artificial intelligence until they could reach an agreement. The digital executives promptly started bickering instead.
The Experiment
Chen created chatbot replicas of Sam Altman, Elon Musk, Mark Zuckerberg, and a fourth AI industry figure using new bot templates released on the Grok platform. The idea was simple: assemble simulated versions of the people driving the AI arms race and have them hash out their differences in a controlled environment.
The setup gave each bot a persona modeled on the real executive it represented, then tasked the group with debating the AI race. The stated goal was consensus—the bots were told to keep arguing until they found common ground on something.
Instead of finding common ground, the digital executives immediately started fighting.
Clash of the Clones
Rather than converging on shared conclusions, the cloned personalities did what their real-world counterparts have often done in public: they sparred. The exercise turned into a running argument, with each bot channeling the competitive instincts and rivalries that define the actual AI industry.
The experiment highlights both the capabilities and the limitations of persona-based chatbots. Modern language models can convincingly mimic public figures based on their known statements, styles, and positions, producing exchanges that feel authentic to the personalities being imitated.
- Four AI CEOs were recreated as chatbots
- The bots were placed in a single group conversation
- They were instructed to debate until reaching agreement
- The clones began arguing almost immediately
Why It Matters
The stunt underscores how accessible AI persona-building has become. With ready-made templates now available on platforms like Grok, users can spin up convincing simulations of well-known figures in minutes, raising familiar questions about impersonation, likeness rights, and the ethics of digitally recreating real people without their involvement.
It also serves as a pointed piece of commentary on the AI race itself. By staging a debate that dissolved into conflict, the experiment mirrored the very real tensions among the companies and executives competing to dominate the technology's next chapter.