Sep 2, 2025 · 47m · a16z
Substack Cofounder on AI Slop Content & the Decline of Social Media
Chris Best · 26m spoken Andrew Chen · 6m spoken Erik Torenberg · 5m spoken Katherine Boyle · 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 this episode of the a16z Podcast, host Eric Torenberg speaks with Substack co-founder Chris Best and a16z General Partners Catherine Boyle and Andrew Chen about Substack's mission to reshape digital media economics. The panel explores free speech, creator ownership, algorithmic feed design, and how direct subscription models counter the rise of AI-generated content.
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 12.6% of the talking time here. How this is scored →
The host as informed peer 2.3 Guest teaching 3.0 Guest disagreement 1.0 The host pushing back 0.6
speaking balance: gold is the host, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 38:29 Chris attacks academic peer review
Chris aggressively rejects conventional academic norms, labeling modern peer review a huge mistake and calling large bodies of published science fake LARPing.
Hardest push from the host ▶ 23:48 Erik presses on ad network risksErik directly challenges Chris's framing on advertising by articulating the bear case where ads induce clickbait and dilute creator incentives.
Biggest teaching moment ▶ 20:39 Chris re-educates on algorithmic designChris reframes the superficial view that algorithms are inherently evil, educating the audience on how objective functions dictate algorithm outcomes.
The host holds their own ▶ 44:36 Erik cites Alex Danko on intellectual supply chainsErik demonstrates strong domain authority by introducing Alex Danko's framework explaining how niche long-form writing feeds broader media translation chains.
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 |
|---|---|---|---|---|---|---|
| Substack's Cultural Impact and Free Speech Commitment | 2 | 2 | 1 | 0 | Erik sets up the discussion on Substack's cultural role, prompting Katherine to recount the 2020 media climate and free speech stance. The host remains in a supportive, facilitating role without challenging the premises. | |
| Building an Economic Engine for Independent Creators | 1 | 3 | 2 | 0 | Chris reframes the premise around free speech, clarifying that while speech is a vital precondition, Substack's core mission is an economic engine for independent culture. Host accepts the distinction. | |
| The Evolution and Restoration of Online Blogging | 0 | 3 | 0 | 0 | Andrew Chen provides a monologue detailing the historical decline of blogging economics and AdSense models. The host is entirely silent during this segment. | |
| Audience Ownership, Right to Exit, and Creative Freedom | 3 | 3 | 1 | 1 | Erik demonstrates good grasp of platform risk and email audience ownership. Chris builds on this by explaining how direct connections give creators power to override platform algorithms. | |
| Substack's Origin Story and Expanding Platform Vision | 1 | 2 | 0 | 0 | Erik prompts Chris on Substack's origin story and evolving platform scope. Chris shares a narrative about his early essay attempt and initial paid newsletter pivot. | |
| Substack Notes, Social Platforms, and Alignment Algorithms | 2 | 4 | 2 | 0 | Erik asks about the transition into Substack Notes. Chris and Andrew educate on platform dependency and reframe anti-algorithm sentiment by explaining objective functions. | |
| Advertising Models, Sponsorships, and Incentive Alignment | 5 | 2 | 1 | 3 | Erik presses Chris on whether launching an ad network risks Substack's core value proposition, providing a clear bull/bear framework. Chris defends alignment over ad models. | |
| AI Technology, AI Slop, and the Scarcity of Great Content | 1 | 3 | 1 | 0 | Chris and Katherine discuss AI slop and creator friction. Chris explains attention scarcity in a post-boredom world while host stays out of the dialogue. | |
| Creator Economics, Unbundling, and the New Media Renaissance | 4 | 3 | 0 | 1 | Erik highlights media price discovery using Noah Smith as an example and asks about unbundling trends. Chris draws an analogy to venture capital liberating software talent. | |
| The Two Futures of Media: Algorithmic Stimulation vs Culture | 2 | 4 | 1 | 0 | Erik prompts on long-term media consumption habits. Chris delivers a philosophical monologue contrasting algorithmic goonbots with intentional cultural participation. | |
| Rethinking Academia, Traditional Publishing, and Knowledge Supply Chains | 4 | 4 | 2 | 1 | Erik introduces adjacent industries like academia, and later demonstrates deep expertise by citing Alex Danko's thesis on intellectual supply chains. Chris takes a contrarian stance on academic peer review. |
Statements from this episode (21)
Chen: Substack saved the dying blogging ecosystem
“This was also an era where, like, the blogging ecosystem was sort of dying. A really important moment to actually save blogging, really, and sort of, like, writing on the internet.”
Boyle: Substack was the only platform protecting free speech
“There was one platform that stood up and said, hey, we are protecting free speech. And that was Substack.”
Boyle: Today's open media culture would not exist without Substack
“Where we are today would not be, like, we would not be where we are today without Substack.”
Best: Free speech is not Substack's primary mission
“I've always seen the free speech thing as sort of an important pillar, but not the main pillar of what Substack is actually setting out to do.”
Best: Liberal media figures bore the brunt of 2020 cancellations
“The people that it really, I think in 20, 20, the thing that surprised me, the people that felt the brunt of that were not conservatives. We're not Republicans necessarily. It was the people in the liberal media. In, you know, in, in, in my telling, I would sa…”
Best: Ousted mainstream journalists boosted Substack's early growth
“The fact that this thing that we were creating this new economic engine for culture that gives you this independence happened to be there at a time where a bunch of the most interesting writers in the world were getting summarily Tossed from their long-time in…”
Chen: Ben Thompson proved subscriptions work, but setup was too painful
“You had Ben Thompson from Stratechery really show that like, oh, there's maybe an alternate model but it was almost like, always like a curiosity and something that was like annoying to actually build.”
Best: Giving users the right to exit forces long-term value creation
“I think the right to exit is really important. People thought that was very dumb. They said, well, if you just let your customers leave, like won't they just leave? And I think in the short run, that might be true, but in the long run, that created the right s…”
Best: Direct audience connections let creators override algorithmic feeds
“And so the direct connection, in addition to being this way, you can bring your audience with you is a way to give humans the power to override the algorithm and say, Hey, I've got this trust relationship with my audience. I want to exercise it and go out on a…”
Best: Paid Substack publications require a free tier to succeed
“If you want to be successful, if you want to make a successful paid sub stack, you have to have a free sub stack.”
Chen: Social platforms are replacing follower graphs with algorithmic feeds
“It turns out then the algorithms are so good that maybe that should be their entire feed is just suggested content. But then what does it mean as a creator to even build a following on one of these platforms? If, you know, even if you have, you know, a 100,000…”
Best: Copying legacy social media ad models would break Substack
“And I think there are definitely like the thing that would break Substack is if we just looked at the same way that the legacy social media things built advertising and said, oh, we're just going to copy that. Like that's going to work because And if we did th…”
Best: Substack plans to build creator sponsorship tools from first principles
“And I think there's a version of unlocking, you know, more economic opportunity, more economic upside that is aligned with the idea of independence, the idea of having differentiated value and quality. And so we're very interested in doing that, but you know, …”
Best: Substack is building a live tool generating multi-format AI media
“We're, we, we're building a live product that basically feels like, you know, I do a, feels like doing a FaceTime call and then magically turns into a highly produced podcast and a YouTube video and a series of short form clips and a transcript. And pretty soo…”
Boyle: Most content on algorithmic 'For You' feeds is now AI slop
“Like, I think there's this horrible meme, like we have, we're, podcasts are over, we have too much content, there's too much online, and it's like, actually, it's the opposite. If you look at any of the For You feeds, I mean, most of it is now AI slot.”
Andreessen: Substack will transform media like VC transformed tech
“This actually reminds me of one of the first things Mark Andreessen ever said to me when we were talking about Substack. He said, you're gonna do to media what the venture capital industry did to software companies or to tech.”
Best: AI and algorithms will supercharge dopamine-driven media
“And I think that that side of media is going to get supercharged. We have very sophisticated AI goon bots now. Is that a good thing? I don't know, but it's happening. And we're going to have that across like, you know, everything, every Short form video, every…”
Best: Peer review is a mistake driving a crisis of fake science
“I think a lot of science is totally broken. I think that they're like, I think that a lot of the, and I think that the scientific project is incredibly important and one of the most valuable things in history, but that the practice of science and the Current s…”
Chen: Selling 10,000 pre-orders makes a book a bestseller
“If you literally just get enough pre-orders that you can get 10,000 units sold, that's like a bestseller.”