Videos » xAI's Grok 5 Is INSANE: Elon Musk's Plan to Reach AGI!

xAI's Grok 5 Is INSANE: Elon Musk's Plan to Reach AGI!

Posted by admin
Elon Musk says Grok 5 has a 10% chance of becoming AGI and could be the smartest AI in the world across every metric. But there's one major problem: Grok 5 still hasn't launched, and none of those predictions comes with independently verified benchmarks. Behind the hype is a massive AI infrastructure project: xAI's Colossus supercomputer in Memphis, reportedly approaching 440,000 GPUs, gigawatt-scale power requirements, and a federal legal battle over gas turbines. Musk has also reportedly discussed a six-trillion-parameter model capable of real-time video understanding and advanced autonomous agents. But can more computing power actually deliver artificial general intelligence? And why do Grok 5's release deadlines keep slipping? We trace five major Grok 5 promises, from Musk's September 2025 AGI prediction and 10% probability estimate to the shifting launch timeline that now places Grok 4.8 and Grok 4.9 ahead of Grok 5. Then we examine what xAI has actually delivered. Grok 4.7 posted improvements in coding benchmarks, including 71.0% on DeepSWE 1.1, 46.3% on CursorBench 4.0, and 37.6% on Terminal-Bench 4.0 according to xAI. With pricing starting at $2 per million input tokens and $6 per million output tokens, Grok has a compelling cost advantage. However, independent comparisons tell a more complicated story. Grok 4.7 trails leading models from OpenAI and Anthropic on several evaluations, while Google's Gemini 4 Argon raises the competitive bar even further. We examine GPT-6 Astra, GPT-6.1 Sol, Claude Fable 5.1, Claude Opus 5.5, and the benchmark differences that marketing headlines often overlook. And there's a revealing twist: shortly before Grok 4.7 launched, Musk himself suggested it would be closer to an older Claude Opus model, while positioning Grok 4.9 as the potential competitor to newer frontier systems. That puts his extraordinary Grok 5 predictions into perspective. We also investigate Colossus 2, reports of a possible 1.1-million-GPU expansion, the NAACP's environmental challenge, and the Justice Department's national-security argument. Plus, we separate existing Grok Imagine, Grok Voice, and Grok Build capabilities from the much harder promises of unified real-time video reasoning and long-running autonomous agents. Finally, we examine the AGI question itself. What would Grok 5 need to demonstrate on independent tests like ARC-AGI and Humanity's Last Exam? Can it handle multi-day projects without supervision? And how would xAI address safety, misinformation, and adversarial content risks inherited from earlier Grok releases? Three outcomes remain possible: a genuine breakthrough, a strong but incremental frontier model, or another release that falls short of enormous expectations. The real test won't be Musk's next prediction. It will be what independent evaluators discover after Grok 5 actually ships. Subscribe to BitBiased for independent AI analysis, benchmark comparisons, and the technology stories behind the headlines. CHAPTERS 00:00 Elon Musk's Grok 5 AGI Prediction 01:44 Five Promises, and the Deadlines That Keep Sliding 05:51 What Grok 4.7 Delivered, and Where It Really Stands 10:41 The Memphis Supercomputer and the Lawsuit Attached to It 13:36 What Real-Time Video and Agents Would Actually Require 14:59 Could Grok 5 Actually Be AGI? 16:54 The Problems Grok 5 Would Bring With It 18:27 Three Ways This Ends, and What to Watch #Grok5 #ElonMusk #xAI #AGI #ArtificialIntelligence
Posted Tue at 9:28 PM
click to rate

Embed  |  106 views