Large companies should empower small, autonomous teams (5-10 people) to experiment rapidly like startups. This "jet ski" model prioritizes speed and validated learning over large budgets and long timelines, de-risking innovation before committing to scale.

The objective of an effective organizational flattening is to streamline the interactions and dependencies between teams, not just to remove people from an org chart. Companies that focus on redesigning workflows and communication patterns first, using frameworks like Team Topologies, achieve sustainable efficiency.

AI tools dramatically reduce the resources needed for idea validation. Leaders should restructure teams by creating small, nimble 'discovery' pods (1-2 people) for rapid idea generation and validation. Successful ideas are then passed to larger, traditional 'execution' teams for scaling and implementation.

If a company creates a siloed "innovation team," it's a sign the main product organization is stuck in "business as usual" maintenance. Innovation should be a mindset embedded across all teams, not an isolated function delegated to a select few.

Using large language models, companies can create 'digital twins' of team members based on their work patterns. This allows managers to run 'what-if' scenarios—testing different team compositions or workflows in a simulation to predict outcomes and flag potential issues before making real-world changes.

Afeyan advises against making breakthrough innovation everyone's responsibility, as it's unsustainable and disruptive to daily jobs. Instead, companies should create a separate group with different motivations, composition, and rewards, focused solely on discontinuous leaps.

Siphoning off cutting-edge work to a separate 'labs' group demotivates core teams and disconnects innovation from those who own the customer. Instead, foster 'innovating teams' by making innovation the responsibility of the core product teams themselves.

Stripe's Experimental Projects Team discovered that embedding its members directly within existing product and infrastructure teams leads to higher success rates. These "embedded projects" are more likely to reach escape velocity and be successfully adopted by the business, contrasting with the common model of an isolated R&D or innovation lab.

To achieve breakthrough innovation, leaders must form a small team and shelter it from the main organization's systems, constraints, and distractions. This isolation provides the mental space required to rethink problems from first principles, rather than being biased by existing structures.