Amazon's AGI team lead, Rohit Prasad, is leaving at the end of 2025; AWS SVP Peter DeSantis will lead a group combining AI, silicon, and quantum computing teams
Todd Bishop / GeekWire :
Context & Ripple Effects
Prasad had been put in charge of a large AI-products organization spanning an Alexa upgrade and other businesses, making this a leadership change in a strategically expanded remit. Amazon had also created a separate agentic-AI group inside AWS earlier in 2025, underscoring how quickly its AI responsibilities have been multiplying.
The move follows another senior Devices and Services departure: Rob Williams’s planned exit late in 2025. Assigning Peter DeSantis a combined AI, silicon, and quantum organization shifts the center of gravity toward an infrastructure-led operating model.
First-order effects
- Prasad’s departure removes the executive who had overseen Amazon’s enlarged AI-products effort; DeSantis becomes accountable for the newly combined AI, silicon, and quantum teams.
- The combination gives one AWS senior leader a single mandate across AI development and the underlying technical domains that support it.
Second-order effects
- AWS will need to clarify how the combined organization interfaces with its existing agentic-AI effort, including the group formed under Swami Sivasubramanian, to avoid overlapping ownership.
- Bringing silicon and AI into the same leadership structure can make infrastructure priorities more directly shape AI product roadmaps and resource allocation.
Third-order effects
- If Amazon sustains this structure, frontier-AI execution may increasingly be organized around control of compute, custom hardware, and platform infrastructure rather than standalone research or product groups.
- The leadership reshuffle also illustrates the organizational strain of scaling AI programs: retaining senior technical leaders and establishing clear boundaries becomes a strategic capability, not merely an HR issue.
The trend: Major cloud providers are consolidating AI software, specialized hardware, and adjacent research functions to turn AI investment into an integrated infrastructure advantage.