- VC Content Club
- Posts
- Sorry Marc, AI is coming for VC, too
Sorry Marc, AI is coming for VC, too
You don’t need to be replaced by AI to become irrelevant.
Paul Smalera
May 06, 2025
Back with a quicker turn than usual to say:
Welcome new subscribers! (A special bonjour to those of you who found me via that French LinkedIn post. Merci!)
The idea for this essay has stuck in my craw for a while now, and I realized I wanted to say something about it, even if it’s a little off-piste for me.
If you are enjoying this newsletter or especially this issue, it’d be great to hear from you or for you to send it to one other person who would appreciate it. That person (hi!) can then mash this button:
On to the essay:
It’s a comforting notion. And a dangerous one.
Because even if AI doesn’t replace human judgment outright, it doesn’t need to. The real threat lies in rendering traditional practices obsolete.
You don’t need to be replaced by AI to become irrelevant. You just need to keep operating like it’s 2015 while the world evolves around you.
And that’s precisely what’s happening in venture capital today.
The Irony Is Hard to Ignore
Venture capital prides itself on funding disruption. Yet, as an industry, it’s been remarkably resistant to it.
Many firms still function like boutique shops with pitch decks:
Basic CRMs.
Disjointed sourcing.
Partner meetings driven by instinct and spreadsheets.
Firms often invest more in swag than in software. While LP letters might get a ChatGPT polish, the systems powering discovery, diligence, and decision-making remain largely untouched.
Why?
Because venture has historically thrived without needing to scale. Returns follow the Power Law, allowing firms to survive on one winner, one reputation, one relationship at a time.
There’s never been pressure to optimize the machine—because the machine was never the point.
Until now.
“There Is No Going Back”
Harry Stebbings aptly noted:
“Venture capital has transitioned from a high-margin boutique business to a low-margin commoditised industry.”
This shift isn’t solely about capital overhang or fund proliferation. It’s about AI gradually consuming the middle of the funnel:
Deal flow increasingly filtered by LLMs before human review.
Competitive analysis a Perplexity query away.
Due diligence accelerated, summarized, and templated.
Investment memos draftable by AI in seconds.
This isn’t the future; it’s already happening. And it’s transforming a human-centered, boutique business into one where speed, synthesis, and story increasingly determine who gets into the best deals first.
Disruption Isn’t About Being Replaced
Clay Christensen’s theory of disruption wasn’t about outright replacement. It was about a change in what the market values—so gradual at first that incumbents barely notice until they’ve lost their edge.
That’s what’s happening in venture.
AI doesn’t need to be better than the best partner at Sequoia to disrupt the industry. It just needs to change the expectations of:
Founders (who now demand faster, clearer feedback and sharper insight),
LPs (who expect data-rich reporting and differentiated theses), and
Platform teams (who are increasingly tasked with content, sourcing, and ops automation—on tooling from 2008).
In other words:
AI isn’t threatening venture by doing its job better. It’s threatening venture by changing the definition of the job.
The Identity Crisis
VC firms have always grappled with an identity crisis. Are they scrappy, boutique operators? Or tech-powered sales machines? High-conviction solo capitalists? Or brand-led institutions with leverageable systems?
Most try to be both. In practice, they often end up as neither.
They hesitate to invest in better infrastructure—CRMs, sourcing platforms, firm-wide knowledge systems—because it doesn’t feel “personal.” They overspend on founder dinners, podcast studios, Patagonia jackets—because it feels like service. And they expect their edge to come from “judgment” while quietly outsourcing their thinking to market consensus and tech media headlines.
That’s not high-margin or boutique. That’s fragile.
To be fair to Marc, his view on this is nuanced:
Andreessen described his job as a nuanced combination of “intangible” skills, including psychological analysis of the entrepreneurs he works with: “A lot of it is psychological analysis, like, ‘Who are these people?’ ‘How do they react under pressure?’ ‘How do you keep them from falling apart?’ ‘How do you keep them from going crazy?’ ‘How do you keep from going crazy yourself?’ You know, you end up being a psychologist half the time.”
“So, it is possible—I don’t want to be definitive—but it’s possible that that is quite literally timeless. And when, you know, when the AI is doing everything else, that may be one of the last remaining fields that people are still doing.”
For sure, being a mentor and startup whisperer to founders will continue to be a valued job. It’s just not clear that it will take being at the prow of an aircraft carrier’s worth of money will be necessary.
Anecdotally, it feels like today’s founders are a more resilient bunch. So many early-stage problems come from struggling to find PMF and making the right early hires and resource allocations. But AI is enabling this crop of startups to stay leaner for far longer than their predecessors were able to.
And a lot of the once-novel “strategy” involved in building a startup has been commoditized such that founders can get a pretty darn good roadmap for their first couple years from Reddit, Hacker News/YC content/YouTube and VC websites themselves.
While this doesn’t mean the Bill Gurley sage-of-startups model is completely obsolete, I think it is something founders are looking for a lot less than they used to. Building startups has simply gotten, at the bottom line, a lot less scary, and thus the cocoon of support around a founder does not have to be as robust as it once was for them to succeed.
Most founders would probably benefit more from talking to an actual psychologist rather than their lead investor!
Mostly Safe = Slowly Fading
So is venture “safe” from AI?
Sure. Just like newspapers were “safe” from blogs. Just like taxis were “safe” from apps.
The real danger isn’t replacement. It’s irrelevance.
The firms that exist today and will still matter a decade from now won’t be the ones with the best slogans or the biggest swag budgets. They’ll be the ones who build institutional memory, synthesize insight faster than a GPT prompt, and make it feel like boutique service—even as they scale like tech companies.
Meanwhile partnerships that are being formed today have a chance to become the a16zs and Sequoias of their generation if they take these lessons (and more) to heart.
The game is changing. The question is:
Will your firm change with it—or just keep telling founders it’s “different” until they stop believing you?
Looking Forward
Have you seen the edges of this shift already—inside your firm or in the market? Are you building new infrastructure? Or are you still hoping judgment alone will save you?
Reply and let me know.