Jan 25, 2023 · 1h 24m · a16z
2023 Big Ideas in Technology (Part 1)
Steph Smith · 25m spoken Jack Soslow · 10m spoken Connie Chan · 10m spoken Anne Lee Skates · 8m spoken Doug McCracken · 8m spoken Sarah Wang · 8m spoken Samit Singh · 5m spoken
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In Part One of a16z's 2023 Big Ideas series, host Steph Smith and six venture partners explore major technological shifts across consumer tech, gaming, enterprise software, and fintech. The discussions highlight emerging trends in AI integration, digital social spaces, social commerce, metaverse identity, and enterprise workflow automation.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. The host holds 32.4% of the talking time here. How this is scored →
The host as informed peer 4.4 Guest teaching 3.6 Guest disagreement 1.3 The host pushing back 3.6
speaking balance: gold is the host, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 42:09 Reframing Creator Bots vs Player Hacks
Jack directly counters the implicit assumption that all bots disrupt fair gameplay by distinguishing game-improving developer bots from rule-breaking player hacks.
Hardest push from the host ▶ 56:36 Challenging the Interoperability PremiseSteph explicitly refuses the assumption that interoperability is imminent, challenging Doug with the counterexample of dominant multi-billion dollar walled-garden games.
Biggest teaching moment ▶ 13:20 Explaining Mobile-Only vs Mobile-First E-CommerceConnie educates Steph on the fundamental infrastructural gap between Western PC-legacy backwards compatibility and Asia's mobile-only commerce innovation.
The host holds their own ▶ 42:33 Citing Deep Blue and Fisher Random ChessSteph displays deep historical knowledge of game theory and AI by citing Kasparov's 1997 match and Fisher Random chess to explain how human game design adapts to AI dominance.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The host as informed peer | Guest teaching | Guest disagreement | The host pushing back | Why |
|---|---|---|---|---|---|---|
| Connie Chan on Social Commerce and Discovery Buying | 5 | 5 | 2 | 4 | Steph challenges the premise that social commerce is new by pointing out existing Instagram ad behavior and comparing it to infomercials and personal mobile-only experiences in Indonesia. Connie educates Steph on structural differences between search-based Western e-commerce and mobile-only live discovery shopping in China. | |
| Anne Lee Skates on Unlocking the Digital 'Third Place' | 4 | 4 | 1 | 3 | Steph explores digital third spaces by comparing physical socialization to digital tools like chess and gaming communities. Anne reframes the concept by explaining psychological safety, global reach for niche interests, and how digital interactions can reverse-translate into physical meetups. | |
| Jack Soslow on Games as a Neverending Turing Test | 6 | 4 | 3 | 4 | Steph demonstrates solid domain knowledge by citing the 1997 Kasparov vs Deep Blue chess match and Fisher Random chess to analyze AI bot integration. Jack gently reframes the dynamic by distinguishing developer-created bots that improve game flow from player hacks that ruin playability. | |
| Doug McCracken on Metaverse Fashion and Digital Identity | 5 | 4 | 1 | 6 | Steph pushes back directly on Doug's interoperability prediction, noting that most major games currently operate in isolated silos. Doug responds by explaining how developer avatar tools like Ready Player Me solve startup engineering costs while meeting Gen Z expectations. | |
| Sarah Wang on Generative AI in Complex Enterprise Workflows | 6 | 4 | 1 | 4 | Steph introduces the concept of fault tolerance in generative AI and invokes Mad Men writers room techniques and Descript feature framing. Sarah elaborates on enterprise workflow integration, proprietary B2B fine-tuning, and comparisons to the shift from on-prem to cloud software. | |
| Samit Singh on LLMs in Fintech and Maintaining Trust | 5 | 4 | 1 | 4 | Steph asks about transparency, regulatory risk, and consumer sentiment when using LLMs in sensitive domains, drawing a comparison to ATM automation. Samit clarifies that back-office SMB workflow automation and customer agent fail-safes are near-term targets where high accuracy is manageable. | |
| Conclusion and Full Ideas List Preview | 0 | 0 | 0 | 0 | Steph Smith provides a solo closing segment summarizing the discussion and teasing part two of the podcast series. As a monologue outro, all power dynamic scores are zeroed. |
Statements from this episode (20)
Chan: Social and video commerce will become mainstream trends in 2023
“So social commerce, discovery commerce, video commerce, these trends are inevitable and 20, 23 will be the year it all becomes obvious.”
Chan: Asian tech trends take one to five years to reach US
“My learning is that there's usually a time lag of at least a year and generally three to five years before some of those concepts that are already mainstream in Asia become very popular here.”
Chan: Infomercial-style video commerce will succeed on mobile phones
“And I think it's gonna work on the phone too. There's no reason why it wouldn't And because platforms are waking up to this and also brands are waking up to this, I think that combination is the important catalyst.”
Chan: Video commerce will spawn a new category of creator-sellers
“And in the same sense, I also think even if we see an increase in commerce for video, you might also see a new category of creators emerge.”
Skates: Americans with a regular local third place dropped post-2019
“In 2019, two-thirds of Americans said they had a favorite local place they went to regularly, but that two-thirds has actually dropped to just a little more than half.”
Skates: Gaming is the best indicator for future consumer technology
“Games is an area that's probably the most advanced and digitally native. And I think in general, it's always a good place to look for the future of consumer technology.”
Soslow: Gamers will not distinguish AI bots from humans within a year
“In the year ahead, you may not know who's who anymore, and you won't mind.”
Soslow: The Turing test is flawed because it tests for deception
“The Turing test, as you say, is like a flawed metric, and the reason for this is in part because It is too low of a bar. It is a binary bar, so it's either intelligent or not intelligent, and it tests for deception, not real understanding.”
Soslow: Future multiplayer games will replace human players with personalized AI
“The ones that I'm most excited about are with like single player games. And then also with multiplayer games where the multiplayer games become personalized, where you actually don't play with other humans anymore. You're playing with perfectly catered AIs.”
Soslow: AI image generation was effectively solved in 2022
“So we saw image pretty much be solved last year and then text continually get better and better.”
Soslow: AI game proliferation will make curation the key creator skill
“I think games is coming for a new renaissance. There's gonna be, Tons and tons of games that are made, but the real skill is in understanding what a good game looks like.”
McCracken: 75% of Gen Z and Gen Alpha will buy digital fashion
“Of that group, two and five already believe that self-expression via fashion is more important in the digital world than the physical, and three and four say they will spend money on digital fashion.”
McCracken: 90% of Gen Z and 94% of Gen Alpha play games
“So first of all, for games, 90% of Gen Z play games, right? And Gen Alpha, it's even higher. I think it's 94% for Gen Alpha.”
McCracken: Over half of Fortnite's $2B annual revenue comes from cosmetics
“Fortnite earned over two billion dollars for multiple years, right? More than half of their revenue comes from cosmetics.”
Wang: Generative AI will drive 100x performance gains in enterprise workflows
“AI will drive 10 to a hundred times performance improvements, showing companies that there is a new way to work, Advancing from text to image to more complex workflows, such as text to SQL queries, or eventually text to Excel modeling and more.”
Wang: AI video tools like Runway can replace 30-year-old incumbent software
“Video editing with companies like Runway really come to mind. You know, this has the potential to replace software that's been around for 30 years.”
Wang: AI apps need end-to-end workflow ownership for long-term defensibility
“And so I think in this next gen of applications built on open AI and open AI's competitors, the way to really differentiate yourself is to further build and develop workflows that are not just sort of a one click button of add AI to X, Y, Z , Is really neat ov…”
Singh: Early fintech LLM adoption will focus on customer support
“I think what we're going to see is, you know, the initial tools that have this sort of level of fault tolerance being around customer engagement tools, for example, in fintech, where both banks and fintechs You know, can use these sort of models to basically a…”
Singh: Distribution remains the key determinant of success despite AI leaps
“However, I think the one thing that founders can forget is that no matter the sort of platform shift, whether it's been internet, cloud, mobile distribution is always key. And so even with this new sort of product leap where maybe, you know, you have consumers…”