Nigeria puts $170.6 million of public money on the table for startups — and hands the whole thing to an outside manager to spend. If you're just joining: regulated-enterprise AI startups have been pulling outsized rounds where governance and compliance are the thing they're selling, not just the packaging. Trase launched on a $107 million seed for highly regulated industries, and Quantifind announced a $200 million growth round led by Summit Partners for AI-native risk intelligence — financial crime, national security. Compliance is the pitch now. This is Startup Fundraising. Today — a sovereign fund-of-funds, a GV space bet with a bank already attached, and a $180 million growth check I want to see the ARR on. Kuramo Capital today, plus LeapXpert and Nebex — but we start with who actually controls Nigeria's money. From ThisDayLive:
Under the arrangement, the federal government will provide an anchor investment of $85.3 million through the iDICE Programme, while Kuramo Capital is expected to mobilise an equivalent amount from private investors, bringing the fund’s minimum capitalisation to $170.6 million.
$170.6 million, and the structure that matters is right there in the fine print: BOI puts in $85.3 million, and Kuramo has to match it dollar-for-dollar from private investors before this thing even hits minimum cap. So the government check forces the private match; it doesn't just soften the risk. And the accountability chain shifts. BOI isn't writing checks to founders — it's writing one check to Wale Adeosun's shop, and Kuramo picks the VC and micro-VC funds that actually deploy. It's a fund-of-funds, so there's a layer between public money and any startup in those 36 states. Right, so squeeze the carry. How much of that $170 million reaches a founder in Kano versus getting eaten by two layers of management fees and two layers of carry? With a fund-of-funds, somebody takes economics twice. And the quieter tell — BOI reaching outside for a manager at all. If Nigeria's domestic VC infrastructure were ready to do this on its own, you wouldn't hand $170 million to an external firm to go find the funds. Here's Mike Wheatley at SiliconANGLE:
Secure business communications startup LeapXpert Inc. said today it has bagged $180 million in a growth round of funding to build out its artificial intelligence capabilities and generate valuable intelligence for enterprises. Riverwood Capital led the round, and Portage Ventures was named as the only other participant.
LeapXpert: $180 million growth round, and here's the part that stands out after the week we've had — Riverwood Capital led it. Named lead, nine figures, and Portage is the only other name on the page. Clean. Growth round, though — I'm hanging on that word. Governed enterprise communications is compliance plumbing. What's the ARR base that turns compliance plumbing into a $180 million check? If they can't name it, this is a Series B in a nicer suit. This is the same regulated-enterprise AI lane as Trase and Quantifind, just with a bigger check. And the compliance gap is real: those informal WhatsApp and Signal threads sitting outside enterprise security produced nine-figure bank fines in the U.S. for exactly that. Fine, the pain is real. But their pitch is: let us capture the messaging your regulators want captured, then sell you intelligence off it. So who validated that product — the enterprises buying it, or someone who isn't also its case study? The Next Web writes:
The space industry can build rockets. Moving money around them is the hard part. A new startup called Nebex wants to fix that, and Google’s venture arm is backing it with $30mn. GV, formerly Google Ventures, led the seed round, Bloomberg reported. Nebex also opened a banking relationship with JP Morgan.
Nebex — $30M seed, GV leading, per Bloomberg. And the detail that actually moves this one: a JP Morgan banking relationship at formation. That's a Tier 1 bank wiring itself in before there's really a market to bank. Right, and I want to know who's on the other end of the first trade. They're building an exchange targeting hundred-million-dollar-plus deals — space firms, foreign governments, investors. Name me the first five counterparties. An exchange with no listings is a website. Structurally, though — GV leading a seed, not tagging along, with JP Morgan as a second institutional anchor. Against a week of club rounds nobody would put their name on, that's a genuinely clean setup. Clean, sure. But the pitch is they make cross-border space deals “large and smooth enough to be worth the effort” — while admitting they don't touch the compliance burden. So the export rules, the tariffs, the national-security checks all stay. What friction are they actually removing? The matchmaking? That's a fee on a spreadsheet. The fee only exists if a $100M deal closes. So the whole model rides on volume they haven't demonstrated yet. GV is underwriting that leap — the belief that this becomes a real market. TechCrunch, with Anna Heim:
Three former DeepMind researchers who created an AI that beat humans at poker have now applied the same technology to trading stocks — and the bet appears to be paying off. Their Prague-based AI lab, EquiLibre Technologies, is now valued at $500 million after raising an undisclosed-sum Series A, TechCrunch learned.
EquiLibre out of Prague: $500 million post-money on a Series A where the round size itself is undisclosed, led by Creandum. And here's what jumps out. Creandum's own VP says it's the largest single check the firm has ever written in one go, per TechCrunch. Okay, but undisclosed round size on a disclosed $500M valuation — that's a company that wants the number that flatters and not the one that dilutes. What's the actual check? Right — we get post-money, we get “biggest ever for Creandum,” we don't get the raise. So we can't back into ownership or how much runway this actually buys. And the thing I'll actually engage with — they already have customers, not some two-years-out roadmap. They're trading billions in daily volume across the S&P and Nasdaq with Tower Research. The reward signal is “how much money did the agent make.” That's the cleanest scoreboard we've had all week. That's the poker DNA — reinforcement learning in a game of incomplete information, now pointed at markets. The revenue proof is live, not projected. That's why Creandum leaned this hard. If Startup Fundraising helps you stay sharp on the funding trail, take a moment to subscribe and leave a review wherever you're listening. It really helps other founders and operators find the show.
You'll find links to everything we covered today in the show notes, so if a funding round, filing, or trend caught your ear, you can dig in from there. That's Startup Fundraising for today. This is a Lantern Podcast.