CZ and Friends

S1E7

Ask.com to AI: Doug Leeds on Curiosity, Persuasion, and Reinventing Leadership

Ask.com to AI: Doug Leeds on Curiosity, Persuasion, and Reinventing Leadership

Released

39 minutes

Photo of Cecilia Ziniti

Doug Leeds

Doug Leeds

Co-Founder of RSL and Former CEO of Ask.com and IAC Publishing

Co-Founder of RSL and Former CEO of Ask.com and IAC Publishing

Dive Deeper

Transcript

Episode Overview

What is a legal product manager? Doug Leeds coined the term on this episode. It names an in-house lawyer who identifies the legal risk and builds the solution into the product itself, the way a product manager would.

It is the skill set that took Leeds from a law firm ultimatum to the CEO seat at Ask.com and IAC Publishing. It runs through this whole conversation about persuasion, psychological safety, and what every tech boom looks like from the inside.

Leeds has one of the more unlikely career arcs in Silicon Valley. He left a law firm after two years, built a company during the dot-com boom, watched it bust, went back in-house, and then spent two decades running consumer internet businesses at billion-dollar scale.

Along the way he competed with Google, put 2,000 employees through improv training, and turned Google's antitrust pressure into Ask's commercial opportunity.

He also goes way back with CZ and Friends host Cecilia Ziniti, co-founder and CEO of GC AI. The two met in 2003, when she was a paralegal at Yahoo and he had arrived through Yahoo's acquisition of Overture, the company that invented sponsored search.

In 2006, he told her he wanted to leave law. She told him she wanted to go to law school.

They have been comparing notes on law, leadership, and technology ever since.

About Doug Leeds

Doug Leeds is the co-founder of RSL (Really Simple Licensing), an open web standard and nonprofit collective rights organization that lets publishers and creators license their content to AI companies, co-founded and co-led with Eckart Walther. He teaches two undergraduate courses at UC Berkeley's Haas School of Business: Leadership by Persuasion and Creativity and Leadership.

A lawyer by training, Leeds spent two years at a law firm before going in-house at Overture (originally GoTo.com), the pioneer of pay-per-click search advertising. He moved to Yahoo after its 2003 acquisition of Overture.

He joined Ask.com in 2006 to run product and became CEO in 2010. He founded and led IAC Publishing, the digital media group whose brands included Dictionary.com and About.com, and later ran EAT Club, the corporate lunch delivery company, through its acquisition by Compass Group.

Key Takeaways

The best in-house lawyers work like legal product managers. They build the fix into the product itself, the way Amazon's Alexa team turned a 50-state wiretap analysis into an on-device listening light.

Authority and influence aren't the same. In-house counsel can dictate the rules but can't make anyone follow them, so the job is getting teams to yes as if it were their own idea.

Psychological safety comes before strategy. Leeds put 2,000 employees through improv training, and teams that can't raise concerns about AI use are the ones with the biggest governance gaps.

Every tech cycle runs the same movie. Leeds has watched an investment rush, a probable bubble, a few winners, and net value creation that outweighs the disruption play out from dot-com to mobile, and he expects AI to follow the same arc.

Content licensing is AI's next infrastructure layer. RSL pairs a machine-readable licensing standard with an ASCAP-style collective so creators get paid when AI trains on their work.

Why Did Doug Leeds Leave Law?

A law firm handed him the decision. Two years in, the firm told Leeds he could choose the firm or his family.

He chose his family, took a policy role at a company that was quickly acquired, and walked away with a severance he describes as five times his annual salary. He used it to start a company, rode the dot-com boom, and watched it bust.

With three daughters at home, his wife made the practical case for dusting off the law degree.

The in-house job he landed at Overture turned out to be the right kind of wrong fit: a legal team full of people who did not particularly want to be lawyers, at a startup where everyone worked outside their job description. Leeds pushed into product and business development.

Then Yahoo acquired the company, and the walls went up.

Leeds recalled the pattern:

"If I did something well, they would say, you did a great job on that. Now we're gonna give it to a business person because a lawyer shouldn't have it now that it's doing well. And if I didn't do well, they would say, this is what happens when you let a lawyer do it."

He calls it the Catch-22 that pushed him out of law for good. The law itself he still loves.

He vacations with his law school friends every year and talks legal issues for fun. His verdict: "It turned out to be a better hobby than a profession."

How Did Ask.com Compete With Google? It Stopped Trying.

Ask.com, fresh off its rebrand from Ask Jeeves and newly acquired by IAC, hired Leeds in 2006 to run product. Ask's product was good; the brand was the problem.

Ask would innovate, ship something new, and watch Google copy it within a week, then collect the credit within two.

Leeds described the dynamic:

"We would innovate. We'd create new things. We release them. And then within a week, Google would copy them."

Google held 90% of the search market and invested billions where Ask invested hundreds of millions. Leeds fielded the same question constantly: how does it feel to compete with a company whose name has become a verb?

His standing answer: "Ask was a verb first." Funny, and beside the point.

By the time he took over as CEO, Ask had spent \$100 million on marketing without moving the needle, and it was not going to.

So Leeds pivoted. He had already started acquiring diverse businesses, Dictionary.com first, which seeded what became IAC Publishing.

For search itself, he made the counterintuitive move. Ask adopted Google's technology and built what he candidly calls an arbitrage business. Buy ads on Google, direct the traffic to Ask's search, sell those users the same ads they would have seen on Google.

At the peak, Ask was spending a billion dollars a year on Google ads, and roughly one in every 10 to 12 Google searches worldwide carried an Ask ad.

The detail most people miss is why Google played along. Google needed a credible, pure-play search competitor to point antitrust regulators toward.

Ask fit the profile perfectly, with real scale and nothing but search. Google's legal exposure became Ask's commercial opportunity, a trade Leeds recognizes instantly in today's headlines.

For in-house lawyers, it is the same lesson in reverse. The regulatory position on the other side of the table is often the opening for your deal.

What Is a Legal Product Manager?

A legal product manager is an in-house lawyer who identifies the legal risk and builds the solution into the product itself, the way a product manager would, instead of delivering a no and leaving the room. Leeds coined the term mid-conversation, and it names the thing he enjoyed most about practicing.

The setup is a structural fact of in-house life. You have a lot of authority over what the legal rules require and almost no authority to make anyone follow them. A mid-level lawyer who walks into a meeting saying "you have to do this" gets escalation and a fight.

Leeds, on the alternative:

"The enjoyment that I had out of being a lawyer was working with teams to figure out a way to get them to do what I needed them to do as if it was their idea. How do we find a way to say yes instead of saying no?"

The canonical example surfaced live on the show. Cecilia was product counsel on Amazon's Alexa, a product most privacy lawyers of the era would have described as a hot mic in your living room.

Her team, alongside Tina Patel, her colleague on Amazon's Alexa privacy team and a later CZ and Friends guest in her own right, mapped wiretap statutes across all 50 states and found the path where the product could exist legally and customers could trust it. The product and engineering teams then built the answer into the hardware: the listening indicator light, plus user controls to delete utterances.

Nobody said no. The compliance answer became a trust feature that drove adoption. It's the same instinct behind why judgment, not a blanket no, is becoming the in-house differentiator.

Why Psychological Safety Comes First

Leeds's answer to "how do you spot opportunities?" comes down to culture. It takes a group of people with enough trust and psychological safety to tell you where the problems are. IAC Publishing itself, he notes, came out of problems surfaced at Ask.

Without that safety, the signal never reaches you, and the pivot never happens, the same risk-reading great GCs build into how they manage growth and crisis.

He engineered that safety deliberately, and the method was silliness. As Ask transitioned into IAC Publishing, Leeds put the entire 2,000-person company through improv training.

His senior executives hated it. Within a couple of weeks, they were telling him it had changed their family lives.

What they practiced was yes-and. It meant active listening and building on someone else's idea.

Leeds explained why it worked:

"Silliness and being able to fail in front of other people in a risk-free way allowed you to build trust so that you could talk about other things."

The experiment is what got him into teaching. At Berkeley Haas, his Creativity and Leadership course is structured entirely around learning through games, and his Leadership by Persuasion course opens with the definition he now leads with everywhere:

"The definition of leadership I use in the class is the ability to get other people to do what you want them to do as if it's their own idea, with interest and desire of their own."

He also teaches a diagnostic question he coined for stalled change efforts. Why aren't people already doing it? Understand the answer, and you know what persuasion has to solve.

Asked what this means for GCs and CEOs trying to bring AI into their organizations, Leeds is direct. The advice is the same. Build the culture where people can say what is going on in their world.

Companies where employees are uncomfortable raising concerns about how AI is being used, sanctioned or not, are the ones with the biggest governance gaps and the biggest liabilities.

Is the AI Boom Different From the Dot-Com Boom?

From the inside, no. Leeds gets the question constantly in the form "should we be excited or scared about AI?" and hears an exact echo of "should we be excited or scared about the internet?"

The pattern he has watched across the dot-com era, mobile, and now AI: a big investment rush, probably a bubble, a few winners, an ecosystem that builds around them, and, on the other side, more value in the world than the disruption destroyed.

His posture is optimism with eyes open, and he expects the AI version to run faster.

Where he sees the disruption most personally is the classroom. AI produces the output teachers used as evidence of student thinking, so the job becomes engaging their intellect through something other than the essay.

His analogy is the calculator: disruption, adjustment, then a tool that handles certain tasks while the teaching moves up a level.

Inside companies, he flags the same gap the improv training was built to surface. Employees are using AI with customer data whether or not there is a sanctioned platform, and every plan the company made now has an AI-shaped hole in it.

Governance, data privacy, and liability are the real problems, and the companies that hear about them early are the ones where people feel safe saying so.

What Is Really Simple Licensing (RSL)?

RSL is Leeds's current act, and it runs straight back through his search career. It is two things: an open web standard that lets any publisher or creator declare, in machine-readable terms, the licensing rules for using their content to train or power AI, and a nonprofit collective rights organization, the RSL Collective, that negotiates those rights at internet scale.

Think robots.txt for the AI era, paired with an ASCAP for internet content.

The logic starts with the deal that built the web. Robots.txt let publishers tell search engines: crawl my content, index it, and send me traffic I can monetize. Leeds, on why that ecosystem broke:

"That is blown up by AI, because they take the content, and they don't give the traffic back, because they're giving you an answer without a link, or a link you would need to use, because the answer is all right there."

RSL launched September 10, 2025, with support from Reddit, Yahoo, People Inc., Internet Brands, Ziff Davis, and Medium, among others. Leeds argues the broken ecosystem hurts AI companies too: models need fresh content, and content stops getting made when creators stop getting paid.

His pitch to publishers and creators is short. It is free to join, and we will give you money.

Recommended Reading

About CZ and Friends

GC AI CEO Cecilia Ziniti talks with the legal leaders rewriting how in-house teams work with AI.

Listen to the full episode

New to Legal AI?

GC AI runs free legal AI classes for in-house teams, from prompting basics to building playbooks, rolling out AI across a legal department, and working with agents.

Get started today.

Get started today.

Get started today.

See how in-house teams run that review — start with a free class, or try the platform on your own work.