
Nov 19, 2026 · 1h 37m
Listen from 29:29
Listen at 29:29
Take-Two bets on human creativity as AI reshapes capital
Anthropic IPO, GTA 6 Launch, Strauss Zelnick Joins | Hemant Taneja
The episode connects two major shifts—AI’s demand for unprecedented infrastructure financing and game development’s push toward larger, longer-lived creative worlds.
- 1Strauss Zelnick says Take-Two pairs strict financial discipline with creative risk, treating franchises as evolving ecosystems rather than release cycles.
- 2Anthropic’s IPO signals that AI companies need new capital stacks combining equity, compute financing, and infrastructure investment.
- 3AI may accelerate production and transform healthcare, but original creativity, regulation, and difficult execution remain limiting factors.
Don't miss
Strauss Zelnick explains why better AI tools may improve game production while leaving original human creativity as the scarce ingredient.
The brief
Strauss Zelnick traces Take-Two’s evolution from a troubled company centered on Grand Theft Auto into a portfolio built around durable creative franchises.
The GTA Online model changed game economics by turning releases into living experiences, while Take-Two’s discipline keeps budgets tight and creative ambition high.
Zelnick argues AI can remove mundane work and expand what developers attempt, but it cannot substitute for the human originality that makes ambitious games matter.
Hemant Taneja frames Anthropic’s IPO as proof that AI growth has outgrown traditional venture assumptions, creating demand for financing that covers compute and infrastructure.
The conversation links AI’s promise in healthcare and biotech to harder questions about regulation, acquisition incentives, cybersecurity, and who finances the machines behind the growth.
What was said on this episode
40 statements · 25 positive · 6 negative · 9 neutral
Video games would become the next major entertainment business.
“the next big entertainment business would be the video game business”
Listen at 7:05
GTA 5 and GTA Online will continue selling and succeeding.
“GTA 5 will continue to sell. That GTA Online based on 5 will continue to be successful”
Listen at 11:16
Take-Two’s success enables it to make very large creative investments.
“the success that we've had, for which we're incredibly grateful, does give us the opportunity. To make these very, very big bets”
Listen at 13:32
Take-Two grants creative freedom within agreed budgets and approaches.
“within the rubric of the agreed upon budget and approach, we then get out of the way creatively”
Listen at 15:30
Grand Theft Auto 6 has the highest video-game score in history.
“The Metacritic score for Grand Theft Auto 6 is 99. It's the highest scoring video game in history.”
Listen at 16:13
Recurring consumer spending represents about 80% of Take-Two’s net bookings.
“recurrent consumer spending, which represents that engagement, is like 80% of our net bookings”
Listen at 17:37
Game businesses will rely less on major releases and more on ongoing engagement.
“The pressure will be reduced on sort of big tempo releases and the engagement will shift to ongoing engagement.”
Listen at 18:27
Mobile will remain the fastest-growing part of interactive entertainment.
“We knew mobile was the fastest growing part of the interactive entertainment business. We felt pretty strongly that would continue to be the case.”
Listen at 22:05
Rockstar teams should take more creative risk.
“I push the teams to take more rather than less creative risk.”
Listen at 25:41
GTA 6 is intended for consumers aged 17 and older.
“this is an M-rated game. This is meant for 17 and above.”
Listen at 26:15
Take-Two carefully restricts its marketing materials to appropriate audiences.
“we're incredibly careful. We want our creative material in the right hands and in no other hands.”
Listen at 31:54
Take-Two’s consumer database includes more than one billion customer records.
“We also have a consumer database across all of our labels that's cross-referenced with more than a billion customer records.”
Listen at 34:10
New technology generally raises productivity and employment rather than reducing employment.
“new technology doesn't tend to reduce employment. It tends to increase productivity, which increases GDP, which actually increases employment.”
Listen at 35:46
AI will not make hit entertainment easy to create, market, finance, or staff.
“making hits, marketing hits, financing hits, employing creative talent, all that's going to remain very, very difficult.”
Listen at 38:27
New creative technologies can expand art and create new art forms.
“art has flourished. Fine art's worth a whole lot more today than it was when cameras were first developed. And an entirely new art form, photography, has also been developed.”
Listen at 39:42
Valuable achievements generally remain difficult despite technological advances.
“the things that are really worth doing tend to be very hard”
Listen at 45:01
Greater technological efficiency will increase creative ambitions.
“as technology enables us to be more efficient and to do better work, Yeah, our ambitions will grow. They will grow.”
Listen at 45:09
Interactive entertainment will undergo major changes in formats and experiences.
“we're going to see an evolution, if not a revolution in the form factor. and the type of entertainment that interactive entertainment brings to consumers.”
Listen at 46:36
Virtual reality would not become a major entertainment medium.
“I was very, very skeptical. And I made sure that our enterprise across our labels was ready for it if I were to be wrong. But I was pretty sure I wasn't going to be wrong. And I wasn't.”
Listen at 48:14
Mobile gaming succeeded because some consumers prefer lightweight experiences.
“I thought it was kind of a flash in the pan because it was too light an experience. And it turns out certain. Consumers want a light experience.”
Listen at 50:10
Effective game streaming will greatly expand console and PC addressable markets within three years.
“within two or three years, we ought to have effective streaming that will massively end. expand our TAM for our console and PC properties”
Listen at 54:23
Take-Two makes strategic decisions on a decade-scale horizon.
“we actually do think in decades at Take-Two”
Listen at 57:22
Anthropic’s rise shows new AI leaders can emerge after an apparent frontrunner.
“the fact that you can have somebody come in that environment and become a leader and come out. That's actually good for the agency in the space.”
Listen at 1:03:01
Other AI companies can also become leaders.
“Makes me think that others can do that again as well. Cause there's existence proof of that.”
Listen at 1:03:10
AI is compressing decades of technological progress into a few years.
“We are just living this incredibly fast movie. We're seeing decades of progress in a couple of years.”
Listen at 1:03:51
The intelligence era will produce substantially larger outcomes than prior venture eras.
“the intelligence era is just going to be, outcomes are going to be that much bigger.”
Listen at 1:06:15
AI companies will require larger, more specialized, innovative financing.
“the capital not only has to scale for us as our founders as well, it also has to become much more nuanced and much more innovative.”
Listen at 1:07:24
Non-equity financing can be fundamentally better than repeatedly selling equity.
“There's other ways to finance your business. It's just fundamentally better.”
Listen at 1:08:21
The distinction between early-stage and growth-stage companies is blurring.
“the line between early and low growth blur”
Listen at 1:09:24
Many structured venture rounds are sloppy financing hacks rather than engineering.
“I wouldn't call that financial engineering. I'd call that sloppy hacks”
Listen at 1:09:47
A small group of companies receives most AI compute and capital.
“70-75% of the compute goes to the same seven or eight companies, and 40-50% of the capital goes to a couple of frontier model companies.”
Listen at 1:12:06
Strong AI startups should demonstrate business evidence with little compute.
“you should be able to produce enough evidence of your business that's interesting with very little compute.”
Listen at 1:13:51
Compute functions more like growth capital when scaling an AI business.
“The compute, to me, is more akin to growth capital in terms of the way it becomes a problem to scale your business.”
Listen at 1:14:10
AI’s major unresolved financing problem concerns compute infrastructure.
“the financial problem really is around how to think about computer infrastructure”
Listen at 1:15:52
AI companies increasingly spend raised capital on compute instead of employees.
“When you raise $100, you used to spend You know, I'd submit probably 70, 80 of that in human beings. Now you have to go spend that in compute”
Listen at 1:15:59
Responsible innovation and strong values can improve company success at scale.
“our belief in responsible innovation that values and scale, doing things the right way actually helps you win.”
Listen at 1:18:00
The largest companies may reach $30–50 trillion market caps within 5–10 years.
“the largest companies are going to be trillion, 10 trillion, and maybe 30 to 50 trillion over the next five to 10 years”
Listen at 1:19:12
Health systems have been among the fastest adopters of AI.
“The health systems have been some of the fastest deployers of that AI.”
Listen at 1:21:51
Regulation and governance must evolve for AI-driven drug development.
“AI-driven targets and drugs needs to kind of rethought.”
Listen at 1:22:38
Biotech markets do not properly incentivize building large independent businesses.
“I actually don't think the incentive structures and markets are set up the right way to go do that.”
Listen at 1:24:30
Statements are attributed to the speaker as said on the episode and reflect their view at the time, not PodLume's. They are not advice.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.