Teresa Carlson, founding CEO of General Catalyst Institute and former AWS Worldwide Public Sector leader, joins host Molly Wood to discuss the transformation of government technology adoption. Carlson built AWS's government business from zero revenue to $10 billion globally over 11 years (2010-2021), literally creating the cloud computing market for governments worldwide.

The conversation covers Carlson's experience navigating the intersection of Silicon Valley innovation and Washington policy, from her early days convincing government agencies to adopt cloud computing to her current role advocating for AI startups in highly regulated industries. Key topics include the parallels between cloud and AI adoption cycles, the importance of landing early government contracts for startup legitimacy, and the critical need for regulatory frameworks in the rapidly evolving AI landscape.

Building the Government Cloud Market from Zero

When Carlson started at AWS in 2010, there was literally 'no way for government to even buy cloud' - no contracting mechanisms, no security frameworks, and no definition of what cloud computing was for federal agencies.

The team started with grassroots education, hosting 'pizza or cookie evenings' in a small Herndon, Virginia building, demonstrating cloud computing to government CTOs and giving them Amazon credits to experiment.

Early resistance came from 'server huggers' - CIOs and system integrators who saw cloud as threatening their domain, requiring AWS to demonstrate that cloud actually gave mission owners more control.

NASA's Mars rover became an early success story: 'We thought it would only rove for four months. It's still roving like eight months later' and using all their data processing capacity - Teresa

The CIA Contract That Changed Everything

AWS won the 2012-2013 CIA contract with just a four-person team, competing against IBM and Microsoft who had 'huge teams' bidding on the classified cloud infrastructure project.

The contract required transforming Amazon's culture - getting people security clearances, building SCIFs (Sensitive Compartmented Information Facilities), and adapting to classified work environments.

'It really was kind of the shot heard around the world because commercial enterprises were like what if the CIA can use cloud we can use cloud' - Teresa on the contract's impact

The CIA chose AWS because they recognized they could build cloud infrastructure but couldn't sustain it due to government contracting limitations that made keeping pace with technology evolution impossible.

Amazon's No-PowerPoint Culture and Leadership Lessons

Amazon banned PowerPoint presentations entirely - 'Everything was a written document. You had to be prolific at writing and telling a story' with crisp, detailed narratives.

'If they ask you a yes or no question, answer yes or no' - Teresa on Amazon's communication style that prioritized efficiency and directness over explanation.

Jeff Bezos emphasized one-way versus two-way doors: one-way decisions couldn't be reversed and required careful consideration, while two-way doors allowed experimentation with manageable risk.

Leaders had to understand P&L completely, acting as general managers who knew 'generally everything about your business' including costs, risks, and ROI calculations.

AI Adoption Outpacing Cloud at Triple Speed

'I used to tell my team at AWS, strap in because you won't see anything like this. And with AI, I'm like, this is a double or triple strap in' - Teresa on AI's unprecedented adoption pace.

Government and regulated industries are adopting AI 'without the same restrictions' that existed for cloud, despite lacking proper definitions, frameworks, or compliance regimes for AI systems.

Carlson predicts an 'optimization point' in AI spending similar to cloud, where initial heavy investment will be followed by companies optimizing token usage and focusing on critical workloads.

The AI landscape will likely mirror cloud's multi-vendor approach: 'You can't change human behavior and human behavior when it comes to buying tech is they're constantly looking' for new solutions.

General Catalyst Institute's Policy Advocacy Mission

The Institute advocates for 1,000+ portfolio companies across highly regulated industries including financial services, healthcare, defense, intelligence, manufacturing, and energy.

'We don't do it on behalf of anyone company' - the Institute works at the industry level, partnering with AWS, Microsoft, Anthropic, and think tanks to shape policy frameworks.

The Institute brings curated groups of founders to Capitol Hill, the White House, and Pentagon, teaching them 'how you tell a story to a lawmaker' and securing high-level meetings.

'We want to be solution providers, not problem identifiers' - the Institute's philosophy focuses on helping lawmakers with constructive solutions rather than just highlighting issues.

Startup Success Stories and Government Legitimacy

Hypocratic AI demonstrated agentic AI patient follow-up calls to senators, with one patient asking 'Are you from Skynet?' but ultimately engaging positively with the AI assistant.

Standard Bots founder Evan Beard 'had never been to Washington DC' but after Institute advocacy now receives calls for feedback and has testified on Capitol Hill twice.

Customer legitimacy is critical for startups: 'What legitimizes you is usage of those tools. And the customer saying it's a good tool' rather than just having great technology.

The current Trump administration is 'the most commercially oriented' Carlson has seen, 'opening up contracting opportunities' and doing 'executive orders that are opening up for these companies.'