Kyle Mack (@KyleTMack) - Twitterプロフィール

固定されたツイート

Kyle Mack@KyleTMack·20 Şub

Yesterday we posted what @MiddeskHQ found when we cross-referenced the @DOGE_HHS dataset against our business identity infrastructure. 300K+ views later, people are asking: what does this fraud actually look like up close? 🧵

Kyle Mack@KyleTMack

HHS just open-sourced the largest Medicaid dataset in history. About $1T in claims data, free for anyone to analyze via @DOGE_HHS. Everyone's looking at what was billed. At @MiddeskHQ, we're looking at who's behind the billing. Here's what we found 🧵

English

5

26

4.9K

Kyle Mack@KyleTMack·24 Şub

Had a great chat with @ReggieCYoung about all things business identity, AI, and the future of @MiddeskHQ

Middesk@MiddeskHQ

AI models aren't the constraint for business identity. Business data is. Middesk CEO Kyle Mack joined Reggie Young on Fintech Layer Cake (presented by Lithic) to talk about why automation in fintech depends on the quality of the identity layer beneath it. Middesk works with roughly 400 government agencies across the U.S. to bring primary-source business data into onboarding and risk workflows. Without complete, reliable data, it's difficult to distinguish a fraudulent entity from a legitimate company that incorporated three days ago. And impossible to automate decisions with confidence. That layer has to be right before anything else can scale. buzzsprout.com/1985266/episod…

English

0

3

386

Kyle Mack@KyleTMack·22 Şub

@typesfast What do you think will be the ways fraud will happen through the broader refund process?

English

0

1

315

Ryan Petersen@typesfast·21 Şub

We're enrolling companies be the first in line to process for tariff refunds start happening. 22 Fortune 500 Companies signed up to get into the process yesterday. We've made the process dead simple and free to enroll. When the window opens, we will be the fastest, most accurate, and most affordable way to get refunds in the world.

English

26

12

193

62.2K

Kyle Mack@KyleTMack·22 Şub

@imthemusic This is only one narrowly defined fraud pattern related to paying out to blacklisted providers

English

0

69

Roseanne Wincek@imthemusic·22 Şub

I am going to get lambasted for this, but.... $1.09T in payments of which ~$2.3B was fraudulent. I know $2.3B is a lot of money, but that means that $1.087T was not fraudulent. So Medicaid had a 0.2% fraud rate. Fraudsters gonna fraud - how is medicaid so good at preventing it?

Kyle Mack@KyleTMack

From 2018–2024, $1.09T in Medicaid payments went to ~1.6M providers. The biggest category was personal care services (in-home visits) at $122B. The fastest growing was behavioral health/substance abuse claims, up 450%+ over 5 years.

English

1

0

4

318

Kyle Mack@KyleTMack·20 Şub

@DOGE_HHS @MiddeskHQ 300K+ views later, people are asking: what does this fraud actually look like up close? x.com/KyleTMack/stat…

Kyle Mack@KyleTMack

Yesterday we posted what @MiddeskHQ found when we cross-referenced the @DOGE_HHS dataset against our business identity infrastructure. 300K+ views later, people are asking: what does this fraud actually look like up close? 🧵

English

1

0

2

1.8K

Kyle Mack@KyleTMack·19 Şub

HHS just open-sourced the largest Medicaid dataset in history. About $1T in claims data, free for anyone to analyze via @DOGE_HHS. Everyone's looking at what was billed. At @MiddeskHQ, we're looking at who's behind the billing. Here's what we found 🧵

English

65

408

1.9K

374.7K

Kyle Mack@KyleTMack·20 Şub

@dpatil @bgurley Couldn’t agree more

English

0

1

33

dj patil@dpatil·20 Şub

Started in Obama with Data.gov and the exec order that all federal data by default should be open and machine readable. Then much of it was put into law during Trump 1. It’s taken far too long to get this out. And needs much more augmentation/cleaning. Good start, far from done

English

1

0

21

1K

Bill Gurley@bgurley·20 Şub

This is so forward thinking. Who pushed it through?

Kyle Mack@KyleTMack

HHS just open-sourced the largest Medicaid dataset in history. About $1T in claims data, free for anyone to analyze via @DOGE_HHS. Everyone's looking at what was billed. At @MiddeskHQ, we're looking at who's behind the billing. Here's what we found 🧵

English

24

74

1.5K

232.9K

Kyle Mack@KyleTMack·20 Şub

At @MiddeskHQ we do this every day for banks and fintechs. The same tools can protect public programs. DM me or reach out at kyle@middesk.com

English

0

3

254

Kyle Mack@KyleTMack·20 Şub

This is one cluster. Across our analysis, we found networks of providers totaling $1.7B in Medicaid payouts linked through shared addresses, officers, and formation patterns. The truth is mapping these connections is difficult, it requires broad data access, tooling, and strong analytical research.

English

1

0

1

274

Kyle Mack@KyleTMack·20 Şub

Yesterday we posted what @MiddeskHQ found when we cross-referenced the @DOGE_HHS dataset against our business identity infrastructure. 300K+ views later, people are asking: what does this fraud actually look like up close? 🧵

Kyle Mack@KyleTMack

HHS just open-sourced the largest Medicaid dataset in history. About $1T in claims data, free for anyone to analyze via @DOGE_HHS. Everyone's looking at what was billed. At @MiddeskHQ, we're looking at who's behind the billing. Here's what we found 🧵

English

5

26

4.9K

Kyle Mack@KyleTMack·20 Şub

@inferencepoint @DOGE_HHS @MiddeskHQ 100% the dive you did is great. Excited to keep following the rabbit hole!

English

0

5

91

7.4K

Samantics@inferencepoint·20 Şub

@KyleTMack @DOGE_HHS @MiddeskHQ There’s so much more to the story x.com/inferencepoint…

Samantics@inferencepoint

DOGE dropped the largest Medicaid dataset in history couple days ago. Everyone's posting screenshots of the top billers. Nobody's digging into WHY the numbers look like that. I downloaded all 227 million rows. Here's what I found underneath. 🧵

English

1

6

86

15.6K

Kyle Mack@KyleTMack·20 Şub

@bgurley Engineers in HHS like @zachterrell57, @nixonandrewg spoke with Axios on making the data public, obviously @elonmusk amplifying and helping to make this possible.

English

0

44

3.4K

Kyle Mack@KyleTMack·20 Şub

@zachterrell57 Excited to dig in! @bgurley this might be your guy

English

0

2

131

Zach@zachterrell57·20 Şub

Awesome Something to keep in mind: with the “cell suppression” policy (dropping rows with small claim counts), we lose ~80% of the data, so the actual total $ amount is much higher Working on ways to lower that cell suppression threshold so we can release a larger version of the dataset

Kyle Mack@KyleTMack

HHS just open-sourced the largest Medicaid dataset in history. About $1T in claims data, free for anyone to analyze via @DOGE_HHS. Everyone's looking at what was billed. At @MiddeskHQ, we're looking at who's behind the billing. Here's what we found 🧵

English

2

1

13

1.6K

Kyle Mack@KyleTMack·19 Şub

TLDR we mapped $1.7B in Medicaid payments to high confidence fraud

Kyle Mack@KyleTMack

HHS just open-sourced the largest Medicaid dataset in history. About $1T in claims data, free for anyone to analyze via @DOGE_HHS. Everyone's looking at what was billed. At @MiddeskHQ, we're looking at who's behind the billing. Here's what we found 🧵

English

2

3

20

2.3K

Kyle Mack@KyleTMack·19 Şub

If you're a journalist, researcher, or data analyst looking at this data we'd love to help. DM me or reach out at kyle@middesk.com Fraud hides in identity. Let's find it.

English

12

38

473

19.6K

Kyle Mack@KyleTMack·19 Şub

This is just a conservative pass of blacklisted providers, revoked licenses, and the businesses directly connected to them. There's much more to dig into: anomalies in average claim sizes, geographic clustering, further validation of individual provider legitimacy.

English

1

12

272

20.6K

Kyle Mack

ディスカバー