The Safeguard Defensibility Read
An independent adversarial read of whether AI in your safeguards workflow has quietly weakened the document, before a lender's desk or an inspection panel finds out for you.
Your resettlement plan, your ESIA chapter, your grievance procedure, your conformance mapping. Each one promises what will always happen and what will never be allowed. Each one will eventually be read by someone whose job is to find where the promise and the practice came apart.
AI is now in that workflow. It drafts the impact section, translates the consultation, summarises the grievance log, maps national law to the lender standard. Most of the time it is faster and quieter than the person it stood in for. The risk lives in the quiet: a claim nobody verified, a community's words flattened in translation, a conformance box ticked against a requirement it does not actually meet, riding the AI into a document whose entire worth is that it survives scrutiny.
And it surfaces downstream, at the lender's desk or the inspection panel, where it costs the most.
The Defensibility Read reads your document the way those readers will, before they do. It is the safeguards edition of the Always/Never Audit, calibrated to IFC Performance Standards, the EBRD Performance Requirements, the Equator Principles, and the people who enforce them.
See the read run on the gold-standard documents themselves: The IFC Performance Standards, audited, and on the World Bank ESS and the Equator Principles in the lender’s safeguards, audited. Grades for enforcement, every quote verified against source.
Convene the council on one claim
Before you read another word about it, watch it work. Type a single always / never claim a safeguards document might make, and a six-seat council of its hardest readers will try to break it in real time. One line, about fifteen seconds, no sign-up.
How it works
- 01Checkout and intakeAfter checkout you receive a private intake page. You send one artifact, up to about fifteen pages: a RAP or LRP section, an ESIA or ESDD chapter, a grievance procedure, or a PS / PR conformance mapping. Plus three lines of context: which standard it answers to, which reader you fear most, and where AI touched the work.
- 02Adversarial council passThe document is read through six hostile-but-lawful seats: the lender, the inspection panel, the affected person, the regulator, the auditor, and counsel. Where the seats converge, you have a structural weakness rather than a matter of taste.
- 03Accountable synthesisThe findings are weighed against the document and signed by one human reviewer with twenty-five years in extractives social performance, resettlement, and the data behind both. The council finds; the human judges.
- 04Memo within five working daysYou receive a written hardening memo, PDF and markdown, within five working days of document receipt and confirmed scope.
A hardening memo for the claims that carry weight.
Strengths come first. A plan that gets the hard things right deserves to know which ones, because those are the commitments worth defending through every revision and every change of operator that follows.
- Which “always” commitments the project can actually deliver, and which are intention written in the present tense.
- Which “never” lines survive a contested project, and which are one cost-pressure away from gone.
- Where AI use has thinned the evidence trail a lender or an inspection panel would ask to see.
- Where the document claims conformance it cannot yet show against PS5, the EBRD PRs, or the Equator Principles.
- The fixes that earn their place first, staged for where the project is now.
Who this is for
Resettlement, land access, and community relations leads carrying a deliverable into lender due diligence. Commission the hostile read before the lender's technical advisor performs it for free, on worse terms.
You produce the document, so you cannot be the one who reads it cold. This is the independent outside pass your own QA structurally cannot give a deliverable it wrote.
A conformance claim to IFC PS5 or the EBRD Performance Requirements is about to be tested by people paid to break it. Find the soft seams while they are still cheap to fix.
The model drafted the impact section, translated the consultation, summarised the grievance log. Find what it smoothed before the smoothing becomes the record.
This is a review. It is not a certification, and it does not put your competence on trial. Your work is what it protects; it is not the thing under inspection. The memo is working input for your team, and no engagement is named by either side without written agreement.
It is also not legal advice. It tells you where a claim breaks against the standard and the reader; your counsel tells you what that means in your jurisdiction.
Founder's launch rate through 2026: $295 USD for one artifact, up to about fifteen pages. The standard rate thereafter is $795.
The deliverable is a written memo, PDF and markdown, within five working days of receipt and confirmed scope. Questions of scope are answered honestly before the clock starts: if your document is a different shape than this read fits, you will be told so and refunded rather than stretched.
Programme-level reviews, full RAP or ESIA engagements, and named-reviewer work are scoped individually through the contact form.
Start the Defensibility Read
Pay securely through Selar. After checkout, you will be directed to the private intake page for document context and delivery notes.
Checkout on Selar - $295 USDThis read runs Rule 4 of the free Interim Protocol on AI in Safeguards Work for you: the hostile read before submission, independent of authorship.