Group 1 - The traditional design process is no longer effective in the AI era, as designers must shift from "delivering solutions" to "assisting execution" [5][6] - AI-driven productivity leaps have rendered the long-standing "research-diverge-converge" process obsolete, as engineers using AI programming tools significantly increase construction speed [6] - The uncertainty of AI outputs makes static interaction simulations less meaningful, necessitating a shift from static canvases to code development environments for design validation [7][8] Group 2 - The core methodology for establishing product credibility has been restructured, focusing on releasing "research preview versions" to gather user feedback on core functionalities [8][9] - The design work is now categorized into "execution landing" and "vision guidance," requiring designers to transition from aesthetic delivery to functional construction and consensus building [10][11] - The vision cycle has shortened from long-term planning (2-10 years) to a more agile timeframe of 3 to 6 months [12] Group 3 - In the AI era, designers are evolving from solution deliverers to execution assistants, with a shift in hiring preferences towards candidates who can incorporate programming tools into their skill set [13] - Managers are encouraged to engage in seemingly low-leverage tasks to maintain a deep understanding of products and teams [13]