Fannie Mae’s new AI-driven crime detection unit: the Pulte/Palantir project

Palantir technologies is less well -known. It describes itself as “bringing the correct data to the people who need it, so that they can make data -driven decisions.” Palantir customers include some of the most mysterious American intelligence services. They do a lot with data, and further, I cannot articulate without revealing my lack of understanding.

So with equal parts of excitement and pre -feel, I see Palantir fighting the mortgage fraud.

The correct data

Palantir brings customers the ‘right data’. So what is that? Simply put, the “right data” for Palantir are everything. Everything means all rotated mortgage loans, from the private data of a borrower to the very public property data. Then there are data from macro industry, originally to securitization. Everything that is, will bump their ability to recognize or prevent fraud.

Correct data – Wrong representation of the occupation:

Let us apply the correct data from Palantir to the problem of mortgage fraud. The lion’s share of fraud that Palantir and Pulte find will be occupancy fraud, so let’s see what the right data in this context means.

Occupation fraud is simple and omnipresent. Borrowers wrongly claim that they will live in the house as a primary stay to be eligible for lower interest rates and payments.

Palantir will scrape public and private data sources for the countless instructions that remain when someone lies about the occupation. For example, they will check rental sites such as Airbnb to see if the property is mentioned there.

Has the borrower registered a car, boat or even a plane at a different address? Palantir will find the addresses for professional licenses and voters registrations. It will scrub the social media to discover where our borrower eats, banks, cleaned dry cleaned. Which ATMs often does the borrower occur? Palantir will see whether the borrower has other properties and will check their value. It will compare square meters and then search for rental advertisements for one of the investment houses of the borrower. Ideally, Palantir will pick up and report the use of electricity use, waste, plus the distance to golf courses or beaches. It should carry out a report on the residential work distance between the borrower and the employer.

At the moment, Fannie’s fraud researchers do most of them by hand. It is not that Palantir will be smarter, but it will be more thorough and quite faster. The CEO of Fannie Mae, Priscilla Almodovar, was quoted in Housingwire on 28 May and said that Palantir found fraud in “10 seconds” (possibly) but “it cost our really talented researchers 60 days” (no). Her explanation is a togable auto-salesman Puffery, but her point is taken: Palantir will be fast.

Correct data – Wrong representation of assets:

Geldwassers use real estate to clean dirt money. That is why the quality of asset documentation must be a primary care of Palantir. There are now commercially available tools to verify the legitimacy of assets.

But Palantir’s data must do much more than just checking whether a document is valid. It should analyze account stardi against borrower profiles, check the herbs of funds and geographical red flags, verify immediately that donors of gifts are related to the borrower (as required by gift letters), and to bump the names of donors against the SDN list.

Finally, with an AI-SUPER tool like this, every recipient of cash-out can be screened in the same way: after all, we do not want our money to be paid to criminals. There is no reason not to analyze every recipient of cash-out.

Asset fraud is like the caramel marble in Ben & Jerry’s salted caramel core – you dig, you will find it. Palantir can be excellent at this.

Follow -up drawings:

When I was a loan bright, I had a borrower against me about occupation. I gave her a cash-out refinancing at home A. But she used that money to buy house B, in which she immediately moved. My QC department caught her and then our legal department ended up in her. Eventually the speed was adjusted and the borrower had to put down more.

The borrower thought it was a small problem, but it all let us float in the wake, in an attempt to resolve a loan that was ‘not of investment quality’.

Finding fraud is so – it is the fast -moving boat whose wake is the static balance of investors, correspondents and insurers.

And what if your findings cause political problems – as discovering that your largest producer is involved? In April 2024, a top -producing loan officer in New Jersey was charged for counterfeit routine information, as pedestrians as HOA Reserve statements.

Finding fraud in the abstract sounds like a just cause. But the reality is that it can be a just pain in the back.

Conclusion:

I checked my DNA up to and including 23andme. I did the Wang-Wab and waited for weeks. I finally got the e -mail – my results were ready. I logged in and got this surprise warning – Stop! You can’t see what you’re going to see!

Oh, yes, right! Do you think you’re Italian? Think again! Have mother’s eyes? Accepted! Heir to the throne? More such as the scuclemere girl.

Just like an unexpected parenthood reveals about Maury Povich, Palantir will discover fraud in your loan files where you would rather have been alienated.

Yes, it will be fast. Yes, it will be efficient. But it will also be heartless and quick.

The actions of Pulte and Palantir, which discover mortgage fraud in all their genetic variations, will demand uncomfortable actions – Require back purchase, recalculation of rates and tense investment interviews.

I am for it – but a word of advice – does not connect this system without deleting yourself. You can be adopted.

Bob Simpson is the CEO of Daylightaml.

This column does not necessarily reflect the opinion of the editorial department of Housingwire and the owners.

To contact the editor who is responsible for this piece: [email protected].