Elon Musk's xAI just secured $6B of funding at a $24B valuation, so what does that mean for $NVDA and the rest of the chip space as money floods AI? @GuyAdami, @TimSeymour, @RiskReversal, and @CourtneyDoming break down whether this wild ride is close to ending: | CNBC Fast Money

Two teams, equal priority. Team A runs small jobs continuously. Team B needs to run one large job. Every time GPUs free up, Team A's jobs fit first. Team B waits indefinitely. The problem: There is no memory of who just ran versus who has been waiting. NVIDIA Run:ai v2.24 introduces time-based fairshare to solve this. The scheduler can now track historical usage. Teams that consumed more recently get lower priority for over-quota allocation. Teams that have been waiting get boosted. Result: prop