Skip to main content
The Diary of a CEO

Uber CEO: At Uber, If You Don’t Perform, You’re Out! Uber Was Losing $3b A Year

103 min episode · 3 min read
·
Dara Khosrowshahi

Episode

103 min

Read time

3 min

Topics

Productivity, Relationships, Investing

AI-Generated Summary

Key Takeaways

  • Transparency as Information Infrastructure: CEOs receive a filtered version of reality shaped by their team's incentives. Khosrowshahi counters this by scheduling regular skip-level meetings with engineers four levels down, who tend to be authority-resistant and candid. He also models radical honesty publicly so employees reciprocate. His core principle: if you want unfiltered truth flowing up, you must first send unfiltered truth flowing down, or the information loop collapses entirely.
  • Turnaround Speed and Talent Replacement: When Khosrowshahi identified Expedia's technology base as broken and leadership as coasting, he replaced nearly the entire senior team rapidly rather than attempting gradual cultural nudging. His decision framework: identify where you'd rather make a mistake. Losing people who can't handle hard truths is preferable to retaining them. At Uber, he explicitly communicates upfront that underperformers will be told directly and exited if they don't improve, filtering for self-selecting high performers.
  • Exponential vs. Linear Thinking in Market Sizing: Conventional market analysis projects linearly, which systematically undervalues transformative companies. Khosrowshahi's team consistently overpaid for acquisitions like Expedia, Hotels.com, and Match.com relative to then-current market valuations, but those prices proved cheap against actual outcomes. The spread between a hockey-stick growth curve and a straight-line projection is where acquisition and investment opportunity lives. Jevons Paradox compounds this: radical convenience or cost reduction expands total market size beyond original category boundaries.
  • Hard Work as a Learnable, Compounding Skill: Khosrowshahi argues that hard work is not innate but a practiced discipline, citing Ronaldo and Michael Jordan as examples where relentlessness differentiates elite from near-elite performers. His personal standard: refuse to be outworked by anyone. At Uber, this is codified in the value "Embrace the Grind." Practically, he checks emails at 5:30 AM and 9:30 PM while protecting 6-8 PM for family dinners, framing flexibility and intensity as compatible rather than opposing forces.
  • AI Disruption Timeline — Intellectual vs. Physical Jobs: Khosrowshahi estimates AI will be capable of replacing 80% of intellectual job functions within roughly ten years, and physical jobs within 15-20 years due to higher capital requirements and regulatory complexity. The critical risk is pace: unlike previous labor transitions from farming to manufacturing, AI compresses the adjustment window below what social retraining systems can absorb. He notes 90% of Uber engineers already use AI tools, with 30% showing measurable productivity acceleration in code output volume.

What It Covers

Uber CEO Dara Khosrowshahi traces his path from fleeing revolutionary Iran at age nine to turning Uber from a $3 billion annual loss into $9.8 billion in free cash flow. He covers hard work as a learnable skill, radical transparency as a leadership tool, AI's disruption of 70-80% of jobs, and autonomous vehicles replacing 9.5 million drivers within 15-20 years.

Key Questions Answered

  • Transparency as Information Infrastructure: CEOs receive a filtered version of reality shaped by their team's incentives. Khosrowshahi counters this by scheduling regular skip-level meetings with engineers four levels down, who tend to be authority-resistant and candid. He also models radical honesty publicly so employees reciprocate. His core principle: if you want unfiltered truth flowing up, you must first send unfiltered truth flowing down, or the information loop collapses entirely.
  • Turnaround Speed and Talent Replacement: When Khosrowshahi identified Expedia's technology base as broken and leadership as coasting, he replaced nearly the entire senior team rapidly rather than attempting gradual cultural nudging. His decision framework: identify where you'd rather make a mistake. Losing people who can't handle hard truths is preferable to retaining them. At Uber, he explicitly communicates upfront that underperformers will be told directly and exited if they don't improve, filtering for self-selecting high performers.
  • Exponential vs. Linear Thinking in Market Sizing: Conventional market analysis projects linearly, which systematically undervalues transformative companies. Khosrowshahi's team consistently overpaid for acquisitions like Expedia, Hotels.com, and Match.com relative to then-current market valuations, but those prices proved cheap against actual outcomes. The spread between a hockey-stick growth curve and a straight-line projection is where acquisition and investment opportunity lives. Jevons Paradox compounds this: radical convenience or cost reduction expands total market size beyond original category boundaries.
  • Hard Work as a Learnable, Compounding Skill: Khosrowshahi argues that hard work is not innate but a practiced discipline, citing Ronaldo and Michael Jordan as examples where relentlessness differentiates elite from near-elite performers. His personal standard: refuse to be outworked by anyone. At Uber, this is codified in the value "Embrace the Grind." Practically, he checks emails at 5:30 AM and 9:30 PM while protecting 6-8 PM for family dinners, framing flexibility and intensity as compatible rather than opposing forces.
  • AI Disruption Timeline — Intellectual vs. Physical Jobs: Khosrowshahi estimates AI will be capable of replacing 80% of intellectual job functions within roughly ten years, and physical jobs within 15-20 years due to higher capital requirements and regulatory complexity. The critical risk is pace: unlike previous labor transitions from farming to manufacturing, AI compresses the adjustment window below what social retraining systems can absorb. He notes 90% of Uber engineers already use AI tools, with 30% showing measurable productivity acceleration in code output volume.
  • Autonomous Vehicles and the 9.5 Million Driver Question: Uber operates 9.5 million drivers and couriers — the world's largest flexible workforce organizer, second only to the Chinese military in headcount. Waymo data already shows autonomous vehicles are statistically safer than human drivers. Khosrowshahi projects the majority of Uber trips fulfilled by autonomous vehicles within 15-20 years. His partial mitigation strategy involves expanding platform work categories — delivery, shopping, and AI model training tasks through Uber AI Solutions — though he acknowledges uncertainty about whether platform expansion will outpace automation velocity.
  • Values Systems Fail Without Specificity and Authenticity: Uber's first post-Kalanick values reset failed because crowdsourced generic terms like "passion" and "teamwork" were forgettable and weaponizable. The second iteration succeeded by anchoring values to behavioral specifics. "Go Get It" signals aggressive execution. "Great Minds Don't Think Alike" signals cognitive diversity. The one value Khosrowshahi wrote personally — "Do the Right Thing, Period" — deliberately omits explanation, placing moral judgment responsibility on individual employees rather than providing a rulebook that can be gamed or exploited.

Notable Moment

Khosrowshahi reveals that a team at Uber built a "Dara AI" — a model trained to simulate his decision-making — so employees could rehearse presentations before bringing them to him. When he asked to see the underlying code, the team refused. He found this both amusing and telling about how AI is already reshaping internal organizational dynamics at Uber.

Know someone who'd find this useful?

Episode Transcript

This year, most business owners have had one thing top of mind, which is how to make AI work for them because its potential is limitless. But if you don't know what you're doing, then you're guessing, which can be incredibly risky. Our sponsor, NetSuite by Oracle, helps businesses to get AI embedded throughout their organization, whether they're earning millions or hundreds of millions. NetSuite is the number one AI cloud financial system, and through their platform, you get all of your accounting, financial management, inventory, and HR in one place. Their AI connector also lets you pick the AI of your choice. Connect it with your business's data, then you can ask it questions, like how much cash on hand have we got in the company, or who are our key customers. And because all of that data is connected, it makes your AI smarter so it can automate routine tasks, deliver specific actionable insights, and help you cut the costs. Already, over 43,000 businesses have chosen to future proof their business with NetSuite. So if you'd like to learn more on how you can help your business, just get their business guide, which is called Demystifying AI, and you can get that free at netsuite.com/bartlett. You come to Uber, you're gonna work your ass off. And if you're not performing, we're gonna let you know. But do you ever worry that they might not be able to deal with the truth? Then they can leave because the most important skill in life is the skill of working hard. And when you see the top athletes, Ronaldo, Michael Jordan, of course they're talented. But the thing that's different about them is they work their asses off. And that's a learned skill. That's not something you're born with. You may be smarter, more talented, etcetera. But I'm not gonna let anyone outwork me. And with that mentality When you joined Uber, it was losing 3,000,000,000 per year. Now it generates 8,500,000,000.0 in free cash flow every year. But it seems that you were forged in such a way that you were gonna be relentless. Yeah. And it really started with being born in Iran. With the Islamic Revolution in 1978, we were not safe there. And I remember at one point, we had these revolutionary guards come into the backyard and bullets went through our living room. So my family came to The US to rebuild their lives. You were eight, eight, nine years old? Yeah. And it really destroyed my dad. Sorry. It's tough for me to talk about it. Alright. Say again. Alright. Let me try again. Seeing that has put me on a road where I just wanted to make my family proud. So I studied bioelectrical engineering, and then my first job was investment banking. And I got to see the process of big companies being built. And then I had the opportunity to take over Expedia. And in your twelve years as CEO, Expedia sales …

Get the full transcript (19,248 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all The Diary of a CEO transcripts →

You just read a 3-minute summary of a 100-minute episode.

Get The Diary of a CEO summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Keep Reading

More from The Diary of a CEO

We summarize every new episode. Want them in your inbox?

Similar Episodes

Related episodes from other podcasts

Explore Related Topics

This podcast is featured in Best Startup Podcasts (2026) — ranked and reviewed with AI summaries.

Read this week's Investing & Markets Podcast Insights — cross-podcast analysis updated weekly.

You're clearly into The Diary of a CEO.

Every Monday, we deliver AI summaries of the latest episodes from The Diary of a CEO and 192+ other podcasts. Free for one show.

Start My Monday Digest

No credit card · Unsubscribe anytime