Descriptions:
Jeff Su delivers a structured breakdown of which AI tools genuinely excel at which tasks, pushing back on the common assumption that ChatGPT, Gemini, and Claude are effectively interchangeable. Drawing on firsthand testing across professional workflows, the video assigns each major model a specific superpower and explains when to lean on it.
ChatGPT is characterized as the most obedient model: given a complex multi-step checklist, it follows every requirement without skipping steps or substituting its own judgment. Su demonstrates this with a hiring rubric containing twelve requirements — ChatGPT satisfied all twelve, while Gemini quietly dropped several. Gemini’s advantage is multimodality at scale: it can natively process video, audio, images, and text together within a one-million-token context window. This makes it uniquely capable of tasks like synthesizing a meeting recording, a 20-slide deck, and a whiteboard photo in a single query — something no other model can currently do natively. Claude’s strength is first-draft quality. Su shows Claude writing a working Go script on the first try for a data export task he could not complete manually, and notes that Claude consistently outperforms alternatives in coding benchmarks and style-matched writing.
The video also covers specialist tools including Perplexity and provides explicit decision logic for model selection by task type. It is part one of a two-part series, with part two covering productivity and creative AI tools including Notion AI and MidJourney.
📺 Source: Jeff Su · Published January 20, 2026
🏷️ Format: Comparison
