Perplexity CEO Aravind Srinivas: Everything is a Distillation of Schmidhuber
Perplexity CEO Aravind Srinivas stated that everything is a distillation of Schmidhuber. This remark points to the inheritance and refinement of modern AI from Jürgen Schmidhuber's early work. The viewpoint reinforces the narrative of AI history, directing funding and attention towards discussions of fundamental research, benefiting academic pioneers from a renewed recognition, while current model companies face pressure regarding originality narratives. Source: Public information
ABAB AI Insight
Aravind Srinivas has conducted research at OpenAI, Google, and DeepMind, later co-founding Perplexity to create an AI answer engine, emphasizing retrieval enhancement and practical reasoning, providing a deep technical perspective on AI development paths. His statement traces the achievements of current large models back to early contributions like recurrent neural networks and meta-learning, aiming to remind the industry to value the sources of fundamental theories while using a historical framework to reduce over-reliance on the myth of originality from a single laboratory. Similar cases can be seen in Yann LeCun's continued homage to deep learning pioneers or Geoffrey Hinton's clarification of the history of backpropagation; current AI is at a stage where engineering breakthroughs coexist with academic origins. Essentially, this represents a transfer of pricing power: the narrative of historical contributions is redistributing the attribution weight for AI progress, with the mechanism positioning current systems as engineered realizations of early ideas through the concept of "distillation," influencing the flow preferences of talent, papers, and capital. ABAB News · Cognitive Laws 1. Today's engineering is often a distillation of yesterday's theory 2. Pioneering work determines subsequent limits 3. Narrative attribution affects resource allocation.