7 Questions We Had Going Into Config Leadership Collective | Figma Blog (opens in new tab)

The post captures lessons from Figma’s Config Leadership Collective, where more than 1,300 design, product, and engineering leaders discussed leading through AI-driven change. Its central argument is that successful leadership depends less on rigid processes or tool expertise and more on adaptability, human-centered design, judgment, collaborative teams, and supported experimentation.

Leading Change Through Experimentation

• Leaders are learning alongside their teams as AI rapidly changes established workflows.
• Rather than adopting fixed processes, they emphasize adaptability and continuous adjustment.
• Executives are experimenting directly with new tools, prototyping ideas, and sharing failures.
• Teams need permission to explore, take risks, and remain enthusiastic even when experiments fail.

Preserving Human-Centered Design Fundamentals

• AI has changed methods, but core principles remain important:
  • Understand users and their workflows.
  • Continue prioritizing craft and quality.
  • Design for people rather than simply following new tools.
• Leaders warn against “chasing the tool” at the expense of human needs and thoughtful design.

Expertise Is Moving Toward Judgment

• AI can increasingly handle execution and task completion.
• Human expertise is becoming more valuable in higher-order activities such as:
  • Taste
  • Discernment
  • Contextual decision-making
  • Evaluating and editing AI-generated work
• Expertise now means selecting the best answer for a particular situation, not simply knowing a single correct answer.

Restructuring Teams for the AI Era

• AI is blurring traditional boundaries between design, product, engineering, and other disciplines.
• Airbnb is organizing work into small, self-contained pods that resemble startups.
• These pods combine core product roles with perspectives such as data science or business expertise.
• Strong editing judgment, diverse viewpoints, and constructive disagreement are treated as essential.
• Effective teams should be scrappy, vocal, ambitious, and willing to challenge one another.

Helping Teams Adopt New Tools

• Adoption requires education, infrastructure, and psychological safety—not just instructions to use AI.
• Expedia is building dedicated support and training to help employees become fluent with AI tools.
• OpenAI recommends starting with small, low-risk tasks instead of imposing large automation programs from the top down.
• A simple use case, such as summarizing a long Slack thread, can demonstrate value and encourage broader adoption.

The practical recommendation is to lead AI adoption as an ongoing learning process: experiment personally, preserve user-centered standards, hire for judgment and curiosity, build cross-functional teams, and introduce tools through manageable, well-supported steps.