DatIA Recaudo
An AI-first expert system for document analysis, payment-capacity assessment, AI-assisted scoring, validation, traceability and compliance. It supports human decisions — it is not an automatic approval engine.
DatIA Recaudo
An AI-first expert system for document analysis, payment-capacity assessment, AI-assisted scoring, validation, traceability and compliance. It supports human decisions — it is not an automatic approval engine.
What needed to be solved
Manually assessing payment capacity and managing a collections portfolio is slow, inconsistent between analysts, and hard to audit as volume grows.
How we tackled it
An expert system that analyzes documentation and produces AI-assisted scoring as an input to a human decision — deliberately not an automatic approval engine: the final credit or collections decision remains a person's call.
System components
Automated document analysis
AI-assisted payment-capacity scoring
Traceability and compliance
Human decision-support dashboard
Credit studies in hours, with every figure backed by its document.
DatIA Recaudo is the credit-study platform of the BankIA family, and almost nobody knows it yet because it is under construction. It reads the applicant's documents (ID, employment certificate, bank statements, tax ID, financial statements), organizes the information, computes capacity to pay and indebtedness with formulas in plain sight, and delivers a study with alerts and a suggested amount and term for the committee. AI prepares; the decision is always a person's. José Santamaría publishes it here with its limits clear, like everything jsadsAI does.
No published price yet: BankIA states it is quoted on launch. It will follow the scale of the whole family, meant to cover maintenance and a small margin, far from what loan-origination systems cost on the market.
Visit DatIA Recaudo ↗From the document to the study, with evidence for every figure
Every extracted figure shows the document and page it came from, with its confidence level. Doubtful data is flagged for an analyst to confirm. The math (income, expenses, indebtedness, capacity to pay, DSCR and a transparent score) is done by code with exact decimal arithmetic, not by the AI model.
The suggested rate is capped at the current usury rate of the Financial Superintendence and each entity configures its own credit policy: AI does not change it. The result is a study the committee can audit, with alerts, missing documents and a recommended amount and term that is never an approval.
It is a platform under construction: the first version is in internal testing with fictitious data, and before use with real applicants it will go through a security review and a legal review of data handling.
- Banks and fintechs that want to speed up application review with an organized file.
- Cooperatives and funds that need consistent studies for the credit committee.
- Large savings groups that run a simple study before lending to someone outside the group.
- Analysts who consolidate documents and formulas by hand today.
What it delivers
As published on bankia.jsadsai.com/soluciones/data-recaudo and dr.jsadsai.com. Under construction; none of this is offered as an operating service yet.
Document list by applicant type
Employed individual, self-employed and legal entity, with mandatory and optional items; the platform flags what is missing.
Reading with evidence
AI extracts each figure and marks it with its source page and confidence level; doubtful data goes to human review.
Rules engine
Capacity to pay, total indebtedness, liquidity, DSCR and score, with a policy configurable per entity and visible formulas.
Usury cap
The suggested rate is limited to the current usury rate of the Financial Superintendence.
Alerts and risks
Expired or missing documents, inconsistencies between supports, an installment above capacity, low-confidence data.
Suggested amount and term
An explainable recommendation for the committee; never an automatic approval.
Printable report and file
Application, result, evidence, calculation, documents with digital fingerprint and decision, with a history of who changed what.
Separate entities
Each entity sees only its own studies and documents, encrypted.
Decision with a note
The analyst approves, adjusts or rejects with a mandatory note. Who decided, when and why is recorded.
See the demo ↗From the document to the study, in six steps
Application
The entity creates the study: individual (employed or self-employed) or legal entity, purpose, amount and term.
Documents
The applicant or the advisor uploads the files from a secure link; they are stored encrypted.
Reading and verification
AI proposes the data with page and confidence; the analyst confirms it against the document.
Calculation
Code computes income, expenses, indebtedness and capacity to pay with visible formulas.
Study
A report with findings, alerts, missing documents and a recommended amount, term and rate.
Decision
The committee or analyst approves, adjusts or rejects. Who, when and why is recorded.
Price: quoted on launch
As published on bankia.jsadsai.com/precios: DatIA Recaudo is under construction and is quoted when it opens. If you review credit every day, join the waiting list and you will be told when testing with entities opens.
DatIA Recaudo
Under construction
Quoted on launch. Same price scale as the rest of the BankIA family.
Join the waiting list ↗Meanwhile: CooperativIA
Support USD 999 / month
Savings and credit cooperatives under your own brand with a validated usury cap.
See prices ↗Meanwhile: NatIA Ahorro
From USD 5.99 / person
Savings groups, already published with a per-person price.
See plans ↗No published price to date. This page is updated when BankIA publishes the plan.
How it keeps AI from getting your money wrong
- AI reads; code calculates. Models extract data. The math is done by code with exact decimal arithmetic.
- Human review of anything doubtful. Every low-confidence figure is flagged for an analyst to confirm.
- The entity's rules. Each entity configures its credit policies; AI does not change them.
- Final human decision. No loan is approved or rejected automatically.
- Designed with the Colombian framework in mind. Law 1581 of 2012, Law 1266 of 2008 and Law 2157 of 2021 guide the design; compliance is not claimed until the legal review.
- It does not approve or reject loans on its own.
- It does not query credit bureaus yet: the scope is the analysis of the documents the applicant provides.
- It does not disburse or move money.
- It does not replace the committee or the entity's regulatory compliance.
- Under construction: first version in internal testing with fictitious data; AI reading is built and still has to be activated and measured for accuracy.
All of DatIA Recaudo, link by link
Every section lives on the project's official site. From here you go straight to it.
What people ask about DatIA Recaudo
DatIA Recaudo is the AI credit-study platform of the BankIA family, created by jsadsAI | José Santamaría: it reads the applicant's documents, computes capacity to pay and indebtedness with verifiable rules and delivers a study with risks and a recommended amount and term for the committee. It is under construction, with a demo at dr.jsadsai.com.
It has no published price: BankIA states it is quoted on launch. It will follow the same price scale as the rest of the family, meant to cover maintenance and a small margin.
No. It prepares the study and recommends; a person decides, with a mandatory note recording who decided, when and why.
Every figure carries its evidence (document and page) and its confidence level; doubtful data is flagged for human review and the calculations are done by code, not by the model.
Not defined yet. Today the scope is the analysis of the documents the applicant provides.
In the entity's space, encrypted and separated from the others. Retention and deletion periods will be defined with the legal review before launch.
System in development within jsadsAI's Risk Intelligence lab.
Services this project demonstrates
These are not related topics: they are the systems actually built in this project, available as services.
Category: BankIA family · AI credit studies. To learn the full scope and how to apply a similar approach to your business, write to us — no strings attached.
Have a similar idea?
Let's design the architecture to turn your idea into a platform, automation or profitable digital asset.
A real human replies. No spam. Free initial diagnostic.