Perplexity CEO Aravind Srinivas Says This Metric Will Define The AI Race Winner

Perplexity AI CEO Aravind Srinivas has predicted how the future will look like and what metric will be used to define who won the AI race. According to him, the firm that generates the most economic value from the energy consumed by artificial intelligence systems will come out of the race as the long-term winner. Talking to CNBC, Srinivas said the most crucial metric for AI companies is 'token value per watt per user,' which is a measure that integrates efficiency, economic output, and performance. As per the Perplexity CEO, the ability to maximise this metric will be more important than making expensive or large AI models. He also said, 'Whoever is able to maximise this particular objective really will, by balancing accuracy, latency, cost,

privacy and intelligence altogether; they’re going to win - that’s what’s going to win long term.'Cloudflare Says AI Traffic Has Surpassed Human Activity On Internet

For those unaware, a token is a basic unit of data processed by an AI model. Every AI prompt is broken down into tokens, each of which requires computing power and energy to process. he also stated that the companies capable of providing the highest value from that energy usage will be in the strongest position over time. He also suggested that high-priced AI models generating strong revenue today may not necessarily translate into a long-term relationship. Srinivas' own firm has been doing a lot of things to optimise the productivity and efficiency of their model. Pretty recently, they launched Personal Computer, which is a tool deciding which AI models should be used for a particular task. Aravind Srinivas also said that the future or AI will involve moving more processing from large data centres to local devices like smartphones and laptops.