AI-Driven Marketer: Master Practical AI Marketing Skills Marketer Reacts to ChatGPT 5.2, Disney+Sora, & OpenAI's Final Move of 2025
Dec 12, 2025
Dan and Travis dive into the surprise release of ChatGPT 5.2, discussing its impressive benchmark scores for real-world tasks. The duo explores OpenAI's partnership with Disney, suggesting it could revolutionize content creation. They also compare the strengths of various AI models, highlighting ChatGPT's superior personalization. The conversation touches on practical use cases, like game assets and coaching, while they speculate on upcoming announcements that could shake up the AI landscape. Tune in for insights and predictions in the race for AI dominance!
AI Snips
Chapters
Transcript
Episode notes
Real-Work Benchmarks Jump Significantly
- ChatGPT 5.2 shows large gains on real-work benchmarks, jumping to ~71% versus human professionals.
- This signals models are moving from narrow tests to handling practical knowledge work more often than before.
I Spy Photo Test Worked On Second Try
- Dan tested ChatGPT vision with an 'I Spy' page and 5.2 found a medium-hard target after prior failures.
- The model located a very hidden stick of gum on the second attempt with a clue, showing practical vision improvement.
Personalization Beats Raw Scores For Users
- ChatGPT's personalization and memory improvements make it more useful day-to-day than raw benchmark leaders.
- Even if other models excel on some tasks, ChatGPT's tailored behavior often wins for repeated personal workflows.