According to monitoring by Beating, Andrej Karpathy, founding member of OpenAI and the originator of the concept of “vibe coding,” today published a post strongly supporting the Claude Code team’s proposal to replace Markdown with HTML. He not only enthusiastically endorsed this shift but also outlined an evolutionary roadmap for AI interaction interfaces, predicting that after multiple iterations, the ultimate output form of large models will be “interactive neural video.” Karpathy believes that the evolution of AI output formats has progressed from the earliest, hard-to-read plain text, to today’s Markdown, and is now gradually adopting HTML—a format with exceptional typographic flexibility—as the new standard. He foresees several intermediate stages (4, 5, 6, etc.) ahead, culminating in the final stage (n): interactive neural video directly generated by diffusion models. To illustrate what this final form might look like, he explicitly referenced the recent no-code pixel-level rendering prototype, Flipbook, released by a former OpenAI researcher. The underlying logic behind this evolution lies in the physical bandwidth of the human brain. Karpathy notes that approximately one-third of the human brain is dedicated to parallel visual signal processing—a “ten-lane highway” for delivering information to the human mind. This determines the optimal solution for human-AI fusion interfaces: the most efficient way for humans to convey instructions (Input) to AI is through speech, while the most effective way for AI to deliver responses (Output) is through high-bandwidth visual content (images, animations, or video). Additionally, he points out that current input methods still have significant limitations; relying solely on voice or text remains insufficient, and there is an urgent need to incorporate spatial referencing capabilities—such as pointing to specific areas on a screen when two people sit side by side. As a practical short-term improvement, he strongly recommends users append “Structure your response in HTML” to the end of their prompts.
Andrej Karpathy predicts AI interaction will evolve into 'interactive neural video'.
MarsBitShare






Andrej Karpathy, co-founder of OpenAI, has shared his vision for the future of AI-human interaction, predicting that "interactive neural video" will become the dominant output format. He argues that visual media aligns better with human brain processing than text. His comments, covered in AI and crypto news, highlight a shift from Markdown to HTML and beyond. Crypto news outlets are closely following his insights as AI and blockchain continue to intersect.
Source:Show original
Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information.
Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.