This page is for any organisation whose money or programme sits behind others: foundations and donors funding grantees, development agencies backing partner organisations, and incubators and accelerators supporting cohorts. We assess AI readiness across the cohort, set the rules, train the teams, and give you one report on where the whole portfolio stands, in fewer delivery days than funding it organisation by organisation. Priced per engagement: you receive a written quote after a free scoping call.
Who this covers. The pattern is the same whether the money is a grant, a programme budget or a place on a cohort: you fund organisations, those organisations are using AI on data you care about, and their capability to use it well is now part of what you are underwriting. Foundations and donors call the organisations grantees; development agencies call them partners; incubators and accelerators call them cohort members. The work reads the same way for all three, and the contract uses your terms rather than ours.
In April 2025 the Center for Effective Philanthropy surveyed 215 foundation leaders and 451 nonprofit leaders. Two of its findings sit uncomfortably beside a third from a separate benchmark of 346 nonprofits.
The capability is being built inside funding organisations faster than it is being passed to the organisations delivering the work. source
Almost two-thirds of foundations and nonprofits report that none or just a few of their staff have a solid understanding of AI and its applications. source
While 92% report using AI, and 81% use it individually without shared workflows. source
If you can answer all eight, you do not need us. If you cannot answer the first three, that is the programme, and the first one takes about four weeks to close.
Self-reported figures understate this, because individual and undocumented use is never counted. Ask the question in writing and compare the answers with what the assessment finds.
Where AI use is individual and undocumented, nobody in the organisation can answer this, which is itself the finding. Across the sector, 81% of nonprofits using AI do so individually without shared workflows. source In your portfolio that data belongs to people your funding exists to protect.
Sector-wide, 47% have none at all, and many that exist are prohibition lists rather than usable rules. source You can ask this question of your portfolio today and the answer takes about a week to gather.
Foundations are adding AI governance questions to due diligence. If a grantee cannot answer, you learn about their paperwork and nothing about their risk.
The risk is concentrated. It sits wherever case data, health data or protection data meets a small organisation with no IT function. That is a short list and it is knowable.
Since 2 February 2025 organisations deploying AI in the EU must take measures to support AI literacy proportionate to role, contractors included. The July 2026 amendment removed the duty to ensure a set level, so the question is now what measures each grantee can show. Funding EU-facing activity can put that duty inside your portfolio.
The sector answer is close to nothing: only about 10% of foundations report supporting grantees on AI implementation. source
Attendance is not evidence. A scored baseline and a re-score are, and they add almost nothing once the assessment is running across the cohort.
Each one is described the same way: where your portfolio is, what you and your grantees hold at the end, how it gets there, and the delivery days. Priced per engagement. You receive a written quote after a free scoping call. Book the scoping call →
You cannot say how many grantees use AI, on which data, under which rules, and the due-diligence questions you are starting to ask return paperwork rather than answers.
One portfolio report: where the cohort sits, which organisations are most exposed, and what to fund first. Each grantee holds its own scored report and has had a 45-minute debrief on it.
The readiness assessment run across the cohort, a debrief per grantee, one report to you. Anonymised or attributed, agreed with grantees up front. About four weeks.
From 13 delivery days, fixed scope. Written quote after a free scoping call.
You know the gap exists across the portfolio and want it closed, with evidence at the end that it was.
Every grantee ends with its own approved AI policy, a working playbook built on its own work, trained staff and a before-and-after score. You hold a closing report showing what moved, organisation by organisation.
Everything in Baseline; a shared policy template, then adaptation clinics so each organisation ends with its own approved policy; two cohort-wide training sessions; a playbook per organisation; a re-score at the end. About sixteen weeks.
From 50 delivery days for up to twelve grantees. Written quote after a free scoping call.
Your grantees are at different stages and will not all be ready in the same quarter, and you do not want to run a programme they have to fit around.
Each grantee gets a Foundation or Adoption engagement when it is ready, with the results those packages carry. You get a quarterly report on what has been drawn and what was found.
You pre-fund a block of individual engagements; grantees draw them down over twelve months; unused balance rolls over or is refunded, as agreed at contract.
A pre-funded block of delivery days (from 25 days), drawn down over twelve months. Written quote after a free scoping call.
It is a shared spine with per-organisation adaptation. Each grantee gets its own policy and its own playbook, built on its own work. It does not get four separate department labs the way a standalone Adoption engagement does.
For most grantees under 100 staff that is the right trade, because the binding constraint is having any approved rules and one usable page. Four department labs can come later.
Where a grantee is larger or carries unusual risk (health data, protection casework, multi-country operations) we will say so in the Baseline report and recommend a standalone engagement for that organisation rather than stretching the cohort format over it.
Ten dimensions covering fluency, data protection, governance, workforce impact, internal politics and psychological safety, with the optional standards mapping in the vocabulary of ISO/IEC 42001, the NIST AI Risk Management Framework and EU AI Act Article 4.
Written as permissions: permitted uses by role, red lines, a data classification table, disclosure rules and an escalation path.
Built from their real tasks. See a sample.
Policy, tool inventory, attendance record and two scores: the file an auditor or their next funder asks to see.
Send the assessment to a single grantee and see what comes back. Or book thirty minutes and we will work through the eight questions against your actual portfolio.