1. Headline & intro
Silicon Valley promised, loudly, that 2020 would change who gets funded. The quiet unraveling of those promises is now impossible to ignore. The latest signal: Kofi Ampadu, the partner who ran Andreessen Horowitz’s Talent x Opportunity (TxO) initiative for underestimated founders, is leaving the firm after the program was put on ice.
This is more than a personnel move. It’s a real‑time stress test of whether diversity‑focused venture programs survive once the press releases stop trending and markets tighten. In this piece, we’ll look at what Ampadu’s departure really says about a16z, about the future of “opportunity funds,” and about how founders who don’t fit the standard pattern should now think about capital.
2. The news in brief
According to reporting by TechCrunch, Kofi Ampadu has informed Andreessen Horowitz staff via email that he is leaving the firm. Ampadu had been leading a16z’s Talent x Opportunity (TxO) fund and program, created in 2020 to back founders who lacked access to traditional Silicon Valley networks.
TechCrunch notes that TxO was paused in November 2025 and most of its staff were laid off. The initiative had been structured around a donor‑advised fund, combining investment capital with network access and support. Ampadu took over leadership from the program’s original head and had been involved for more than four years. After TxO’s pause, he appears to have worked with a16z’s newer accelerator, Speedrun.
The last TxO cohort ran in March 2025. TechCrunch reports that Ampadu’s exit likely marks the end of the TxO chapter, as many major tech firms scale back or reshape earlier commitments to diversity, equity and inclusion.
3. Why this matters
At first glance, one partner leaving one firm might feel like inside‑baseball venture gossip. It isn’t. Ampadu’s exit is symbolic of how quickly the industry is de‑prioritising structured support for founders who don’t look or sound like the typical Sand Hill Road archetype.
TxO was one of the highest‑profile attempts by a megafund to institutionalise a different on‑ramp into venture: not through elite universities and warm introductions, but through an explicit mandate to source “out‑of‑network” talent. When the person leading that mandate exits shortly after the program is paused, the message to underrepresented founders is clear: don’t count on large generalist funds to treat inclusion as a durable strategy. It is, at best, a cyclical initiative.
Who benefits? In the short term, a16z gets to simplify its narrative and resource allocation around its core businesses: massive thematic funds, crypto, games, AI, and its Speedrun accelerator. Who loses are exactly the people TxO was designed for: talented operators without pedigree or proximity. These founders now face an even more binary choice between bootstrapping and chasing a very small set of niche diversity‑focused vehicles.
It also matters to LPs and corporates. TxO was a test case for whether philanthropic‑style structures (via a donor‑advised fund) could be a scalable way to fund “impactful” startups within a top‑tier VC brand. The implicit verdict, at least at a16z, appears to be: not compelling enough to survive the current cycle.
4. The bigger picture
To understand this move, you have to zoom out to 2020. In the wake of George Floyd’s murder and global protests, U.S. tech and finance raced to launch opportunity funds, diverse founder accelerators, and headline‑grabbing commitments. Venture firms set up side vehicles for Black and brown founders; corporates promised billions in new funding and procurement; every conference panel suddenly had “equity” in the title.
By 2023–2024, the backlash had started. Public markets soured on anything with an ESG label. A wave of legal and political challenges in the U.S. targeted race‑conscious programs, from university admissions to private grant funds like Fearless Fund. Tech companies under budget pressure quietly reduced DEI headcount and marketing, while telling investors they were “refocusing on core priorities.”
Within that arc, TxO’s pause and Ampadu’s departure don’t look like anomalies; they look like part of a pattern. Dedicated vehicles for underrepresented founders are often the first to be questioned when the macro environment turns. They are framed as nice‑to‑have, even though the main funds that remain continue to over‑index on the same insider networks that produced the last generation of unicorns – and the last generation of bubbles.
Compared with competitors, a16z at least built a branded initiative with some runway and experimentation. Many other top‑tier firms limited their response to a few high‑profile hires or one‑off checks. But the end state is converging: diversity is being reabsorbed into generic “we’re open to great founders from anywhere” messaging, without specific structures that would actually change dealflow.
5. The European / regional angle
It might be tempting for European founders to dismiss this as U.S. culture‑war collateral damage. That would be a mistake. Silicon Valley still sets the narrative for global venture. When its biggest brands quietly roll back specialised diversity initiatives, it becomes harder for emerging managers, corporate funds and even public‑sector programmes in Europe to justify their own more targeted approaches.
European VC already has a concentration problem: capital flows heavily to a few hubs (London, Berlin, Paris) and a narrow demographic of founders. Data from multiple studies over the past years show that all‑female or mixed teams still capture only a small single‑digit share of total VC in Europe. For Black founders in Europe, the numbers are even worse.
The a16z story also intersects with regulation. EU institutions talk a lot about inclusion – from the Digital Markets Act to cohesion funds and various innovation schemes – but most of that machinery is indirect. There are comparatively few large, private, name‑brand funds in Europe that have made a TxO‑style bet on systematically backing “out‑of‑network” founders. Instead, the heavy lifting is often done by smaller impact funds, public co‑investment vehicles, or corporate accelerators.
If the U.S. retreats from this space, European policymakers may feel more pressure to pick up the slack – or, less optimistically, may decide the political risk is too high to double down.
6. Looking ahead
What happens next depends on where you sit.
For large venture firms, Ampadu’s exit is a data point they’ll quietly study. Was TxO perceived internally as mission‑aligned but economically marginal? Did the donor‑advised structure create governance or incentive friction? Expect other firms to use this as an argument for folding any “inclusion work” into existing funds, rather than maintaining separate branded initiatives that can later be judged as underperforming.
For founders, the practical lesson is starker: build your fundraising strategy assuming that most big‑name VCs will not operate dedicated access programs for long. If they exist, treat them as opportunistic bonuses, not core pillars. Instead, focus on:
- Specialist funds that have inclusion baked into their mandate and governance (often smaller, but more aligned).
- Ecosystem programs run by corporates, cities, or universities that are less sensitive to short‑term market cycles.
- Cross‑border capital: European, African, or LatAm funds that explicitly look for “overlooked” segments.
Over the next 12–24 months, watch for three signals: whether a16z formally sunsets TxO; whether Speedrun quietly absorbs any of its philosophy; and whether Ampadu resurfaces with a new vehicle or platform of his own. If he does, that will say more about where the next generation of inclusive capital is likely to live: outside the walls of the mega‑funds.
7. The bottom line
Ampadu’s departure from a16z is less about one partner and more about the industry’s appetite for structural change. When the most vocal champions of diverse founders inside top‑tier firms move on, it exposes how fragile those commitments were. The question for founders and LPs alike is simple: will inclusive venture be rebuilt as a core investing thesis, or relegated to a short‑lived post‑2020 experiment that the industry is already trying to forget?
