AI NEWS: OpenAI and xAI spark the AGI race, AI detects consciousness and Google's "AI Money"

Quick Overview

The race for AGI is heating up with OpenAI and xAI announcing massive compute scale plans, Sam Altman aiming for a factory producing one gigawatt of new AI infrastructure weekly, and discussions covering AI's disproportionate benefits for people with ADHD, the potential for agentic money protocols involving Google and Coinbase, and AI detecting consciousness through micro-movements in comatose patients.

Key Points: LLMs like ChatGPT provide disproportionately more benefits to people with ADHD, increasing their productivity, organization, and job satisfaction by reducing cognitive load. The AGI race is intensifying as Sam Altman announced a vision for a factory to produce "a gigawatt of new AI infrastructure every week," while Elon Musk launched Colossus 2, a 1.1 gigawatt supercluster. Sam Altman's energy investment strategy includes Helon Energy (nuclear fusion), Exowatt (solar-powered heat storage for data centers), and Oaklo (micro nuclear reactors), suggesting energy is the real growth market. A new computer vision model named 'CME' analyzes facial micro-movements and subtle skin color changes in comatose patients to detect underlying consciousness before visible signs appear. Google, Coinbase, and 60 partners unveiled an agentic payment protocol (A2A) and marketplaces, aiming to facilitate seamless micro-transactions between autonomous agents for services like booking events. The discussion touched upon life extension research where AI modeled macrophage types, suggesting that removing one type accumulating with age could significantly slow the aging process. Zoom introduced AI avatars that can attend meetings, reference user documents, and summarize content, though the speakers questioned the necessity of the visual likeness aspect over a standard agent.

Context: This informal podcast episode features two hosts discussing recent major developments across the artificial intelligence landscape, including competitive scaling by major players, novel applications of AI in medicine and finance, and personal reflections on neurodiversity. The conversation centers on the escalating requirements for computational power (compute) needed to advance AGI, touching upon the geopolitical implications and energy demands of this scale-up, alongside tangents about personal experiences with ADHD and the implications of new AI tools.