The Nonprofit world has prediction problem

 

…This founder thinks he has the fix, if someone will help him build it

 

Semaform AI wants to tell donors whether their money will actually work before they send it. Right now, it exists mostly in one founder’s head, a pitch deck, and a very full contacts list from the World Economic Forum, the UN, CES, Cannes Lions, and Afrotech.

Philanthropy runs on an honor system that would embarrass a used car lot.

A donor reads a mission statement. Skims a budget. Maybe glances at a Charity Navigator score last updated during a different presidential administration. Then they write the check, close the laptop, and wait. Six months later a glossy PDF arrives explaining what happened to their money, in the same way a doctor might explain what happened to a patient after the funeral.

Nobody in venture capital would fund this way. Nobody in public markets would trade this way. Somehow it is exactly how the world moves trillions of dollars toward the problems it claims to care most about.

Chaste Inegbedion has a name for the fix. He calls it Semaform AI, and he will be the first to tell you it does not exist yet.

“We are pre-MVP,” he said. “I say that on purpose, out loud, in rooms full of people who could write the check that changes that sentence.”

That is either the most honest sentence a founder has said on the record this year, or the smartest one. Possibly both.

The pitch, in plain English

Semaform AI is trying to turn giving from a leap of faith into something closer to a weather forecast. A donor or foundation logs their focus areas, their giving history, their risk appetite. The system studies nonprofits and initiatives against that profile, then runs a predictive model on likely outcomes before a dollar moves, the same way an investor runs diligence before a term sheet.

Once the money is out the door, the system does not go quiet. It keeps watching. A road project outside Kabul. A bridge in Kandahar. A cost per kilometer that suddenly looks wrong. A GPS ping that lands somewhere it should not. Semaform AI is built to flag the anomaly while the concrete is still wet, not eighteen months later in a footnote.

“Traditional philanthropy treats data like an autopsy,” Inegbedion said. “We are trying to build something closer to a navigation system. You do not want to find out you took a wrong turn after you have already arrived somewhere terrible.”

It is a good line. It is also, notably, a line from a man whose own platform is still a pitch deck and a very ambitious roadmap. Semaform AI has the diagnosis nailed. What it does not have yet is a working product, and that is exactly the point of telling this story now instead of after the ribbon cutting.

The part where the founder’s resume does something unusual

Most pre-MVP founders walk into a pitch meeting with a deck and a dream. Inegbedion walks in with a guest list.

Through Semaform Foundation’s Handshake Community, he has spent years convening founders, funders, and policymakers on the sidelines of the World Economic Forum, the United Nations, CES, Cannes Lions, and Afrotech. Not attending. Convening. There is a difference, and it shows up in the caliber of the room.

At the World Bank Spring Meetings, the Handshake Summit and Awards centered on relationship capital, described from the stage as the art of transforming human connections into measurable economic outcomes. The theme, The Offline Survival Kit, Building and Monetising Impactful Communities, opened with a keynote from Oluseun Onigbinde, founder of BudgIT and chair of the Handshake Summit and Awards. The evening’s Technology Showcase featured ConcordeApp and Semaform Foundation alongside Utiva and CrossKudi, ahead of an awards ceremony honoring the Spring 2026 Super Seven.

Weeks earlier, at the 70th UN Commission on the Status of Women, Inegbedion took the moderator’s chair himself, leading a high level panel called AI Driven Climate Leadership and Deal Making for Women’s Prosperity as part of the Handshake Summit’s AI and Impact Showcase. The panel brought together leaders from ServiceNow, M&T Bank, the World Energy Council, and beyond to dig into how digital trust infrastructure can unlock capital for women-led initiatives, the same core question Semaform AI is now trying to answer with a product instead of a panel.

So, the usual pre-MVP question, who is actually going to use this thing, has an unusually concrete answer. The Rolodex was built first. The product is catching up.

“I have spent years building the room,” Inegbedion said. “Now I need the tool that proves the room was right to show up.”

Where anthropic fits, and why this is not a stretch

Here is the part that should make an editor’s ears perk up, because it is not a hypothetical. It is a live program.

Anthropic already runs Claude for Nonprofits, giving mission driven organizations discounted access to Claude along with built in connections to the tools nonprofits already run on, like Blackbaud and Candid, plus training built for teams that have never had a dedicated engineer. That alone would move Semaform AI faster than most bootstrapped nonprofits ever move.

The bigger swing is Claude Corps, the $150 million fellowship Anthropic launched with CodePath and Social Finance. It places 1,000 early career, Claude trained fellows inside mission driven organizations for a full paid year. CodePath employs them and runs the training. Social Finance measures whether it is working. Anthropic funds it and provides the Claude expertise.

Every pre-MVP nonprofit hits the same wall eventually. The vision is real, the need is real, and the organisation still cannot afford to hire the engineer who would make either of those things matter. Claude Corps was built to knock down exactly that wall, for organizations that would otherwise spend years fundraising just to hire their first technical hire.

“Semaform AI is not asking anyone to believe in magic,” Inegbedion said. “It is asking for the same thing. Current AI proved works when public institutions and private capability actually sit at the same table. Give the room its engineer, and the room builds the thing it already knows how to describe.”

The pipeline that already exists

Semaform Foundation does not operate in isolation. It previously partnered with BudgIT and the Open Gov Hub, the Washington DC coworking home to a cluster of nonprofits and civic technology organizations already building in the transparency, accountability, and governance space. That matters more than it sounds like it should.

A predictive philanthropy engine is only as good as the pipeline of real projects and real organizations feeding it. Semaform AI does not need to go recruit that pipeline from scratch. That relationship already exists. Open Gov Hub is full of exactly the kind of small, mission driven, chronically under resourced organizations that a tool like this was built to serve first, the ones least likely to ever land a Bridgespan engagement or a McKinsey pro bono team, and most likely to benefit from a system that tells a funder their work is worth the risk before the funder has to take it on faith.

That is the version of this story that should matter most to the institutions with the biggest checkbooks. The Clinton Foundation, the Gates Foundation, the George Kaiser Family Foundation, and the family offices that move real capital into this space every year are not short on funding to deploy. What they are short on is a reliable way to know, in real time, whether that capital is landing the way they hoped, and a way to extend their impact past the size of their own check.

That is a different ask than a grant. A foundation writing a check funds one project. A foundation backing Semaform AI funds the infrastructure that helps every project in that pipeline get funded smarter, monitored better, and course corrected faster, starting with the network Semaform Foundation already built through BudgIT and Open Gov Hub and extending to every nonprofit those institutions already fund.

“Foundations like these do not need more places to put money,” Inegbedion said. “They need a way to make every dollar they have already committed go further, and a way to find the next fifty organisations worth funding before those organisations even know how to ask. That is what throwing real weight behind Semaform AI looks like. Not a grant line item. A force multiplier.”

Semaform AI does not have a product yet. It has a founder who has already put a UN summit, a World Bank showcase, and CES into the same conversation about how philanthropy should work, and an existing network of mission driven organisations, built through BudgIT and Open Gov Hub, ready to be the first proof points the moment the engine is built. Betting against him closing this next gap would require ignoring everything that got him this far.

 

. Inegbedion is a technology entrepreneur, journalist, and social innovator. He is the Head of Happiness at ConcordeApp, an AI platform transforming how people build meaningful professional relationships, and the Head of Failure & Social Experiments at Semaform Foundation, a nonprofit advancing technology for social impact and health equity. Chaste serves as Global Editor of DA News and is a regular contributor to BellaNaija, BusinessDay, and Premium Times, where he writes on artificial intelligence, innovation, entrepreneurship, and global development. His reporting has covered major international forums, including the World Economic Forum, the United Nations, CES, Cannes Lions, Afrotech, the World Bank, and AI for Good.