According to The Rundown AI on X, Sabi emerged from stealth with a noninvasive brain–computer interface beanie embedding 70,000 to 100,000 miniature EEG sensors, enabling text input by imagining words, with a first product targeted for late 2026 and a baseball cap variant to follow. As reported by The Rundown AI, Sabi has collected 100,000 hours of brain data from 100 volunteers to train a brain foundation model, positioning the system for generalizable decoding without surgery. According to The Rundown AI, investor Vinod Khosla, an early backer of OpenAI, argues mass-market BCI must be noninvasive to reach billions of users, underscoring consumer-form-factor design as a go-to-market strategy. For AI businesses, the opportunity lies in foundation-model-powered neural decoding, edge inference on wearable EEG arrays, and new input modalities for AI assistants and productivity apps, according to The Rundown AI.

According to OpenAI (@OpenAI), during a recent podcast episode, OpenAI CFO Sarah Friar and Khosla Ventures founder Vinod Khosla discussed with host Andrew Mayne the critical shortage of compute resources in the AI sector. They emphasized that compute is currently the scarcest resource driving AI advancements, with business and research demand continuing to outpace supply. The conversation highlighted that solving the compute bottleneck is essential for scaling AI applications and democratizing access to AI benefits across industries. Both leaders pointed out that increased investment in infrastructure and innovative hardware solutions are necessary to meet escalating market needs and unlock new AI business opportunities. Source: OpenAI (@OpenAI), January 20, 2026.