Just hours after OpenAI unveiled GPT-5, Elon Musk took to X (formerly Twitter) to assert that his company xAI’s flagship model, Grok 4 Heavy, outperforms GPT-5. Musk proclaimed, “Grok 4 Heavy was smarter 2 weeks ago than GPT-5 is now and G4H is already a lot better”
He also touted Grok 4’s top-tier performance in benchmarks, citing its lead over GPT-5 in abstract reasoning tests like ARC-AGI
Further stoking competition, Musk teased Grok 5, saying it will be out before the end of this year and will be “crushingly good”
What OpenAI Says—and Industry Reactions
OpenAI officially described GPT-5 as a “major upgrade”—highlighting improvements in coding, reasoning, creative writing, and visual perception, and noting its availability to all ChatGPT users
While GPT-5 gained praise for its accessibility and cost-effective rollout, some analysts called its progress “evolutionary rather than revolutionary”, with reports that Grok 4 Heavy outperformed it in select knowledge and reasoning benchmarks
Why This Matters
- Escalating AI Rivalry: Musk’s public challenge amps up competition between xAI and OpenAI, with Microsoft also deeply involved via its integration of GPT-5 across platforms like Azure and Copilot
- Benchmark Prestige: Musk’s reference to ARC-AGI performance heightens the stakes on AI reasoning capabilities.
- Next-gen Anticipation: The promise of Grok 5 by year-end sets expectations for another AI leap.
