IP Library › Granted Patent US 12,659,332
Granted Patent B2
US 12,659,332 · App. 18/633,144 · Granted Jun 16, 2026

Secure platform for processing data

Inventors: Yannick Lecroart (La Grand Motte, FR); Kamal Ayoubi (Montpellier, FR); Loïc Maisonnasse (Mauguio, FR)
Assignee: ATOS FRANCE
H04L63/1425H04L41/16H04L63/1458
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,659,332
App. No.
18/633,144
Granted
Jun 16, 2026
Kind
B2
Abstract

The invention relates to a data processing platform that includes a computing server and a compliance server distinct from each other. The computing server receives a data stream from at least one source, and executes, based on the data stream, a data processing process including at least one task. At each execution of a task, the computing server transmits, to the compliance server, a request including contextual data representative of an execution context of the task. The compliance server loads, from a blockchain, pre-established compliance data; compares the received contextual data with the loaded compliance data; and generates an alert based on a result of the comparison.

Claims (35)

1 . A data processing platform comprising:

an input port that receives a data stream from at least one source through a communication network,

a computing server; and

a compliance server;

wherein the computing server and the compliance server are distinct from each other and in communication with each other,

wherein the computing server is connected to the input port and wherein the compliance server is not connected to the input port,

wherein the computing server comprises one or more processors operably connected to a memory and is configured to

store at least one artificial intelligence model, said at least one artificial intelligence model being previously trained or not,

execute, based on the data stream that is received, a data processing process comprising at least one task, said at least one task being at least one of

training the at least one artificial intelligence model based on training data dependent on the data stream that is received; and

in case the at least one artificial intelligence model is previously trained, processing by the at least one artificial intelligence model that is previously trained, raw data dependent on the data stream; and

transmit to the compliance server, upon each execution of a task of the at least one task of the data processing process, a request comprising contextual data representative of an execution context of the task, wherein said contextual data comprises a description of input data based on which the task was executed, said input data comprising training data used to train an artificial intelligence model, and/or parameters of the artificial intelligence model,

wherein the compliance server comprises one or more processors operably connected to a memory and is configured to

receive from the computing server the request comprising the contextual data;

communicate with a blockchain, said blockchain being external to the data processing platform, and load, from said blockchain, compliance data that is pre-established; said compliance data being associated with an expiration date;

compare the contextual data with the compliance data that is loaded from the blockchain only if a current date is earlier than the expiration date; and

generate an alert based on a result of said compare, in case of a mismatch between the contextual data and the compliance data.

2 . The data processing platform according to claim 1 , wherein the compliance server is configured to command, in case of a mismatch between the contextual data received from the computing server and the compliance data that is loaded, an interruption of execution of the data processing process by the computing server.

3 . The data processing platform according to claim 1 , the compliance server is configured to record, in the blockchain, compliance information representative of a situation where, for each of the at least one task of the data processing process, the compliance server has determined a match between the contextual data corresponding therewith and the compliance data that is loaded.

4 . The data processing platform according to claim 3 , wherein the compliance information comprises one or more of a timestamp and an electronic signature from the compliance server.

5 . The data processing platform according to claim 1 , wherein the compliance data comprises, for each task of the at least one task, a description comprising one or more of

acceptable input data for executing the each task;

acceptable commands for executing the each task;

rules relating to acceptable actions during execution of the each task;

acceptable steps for executing the each task;

authorized actors to execute the each task.

6 . The data processing platform according to claim 1 , wherein, for each task of the at least one task, the contextual data further comprises a description comprising one or more of

commands implemented during execution of the each task;

actions performed during execution of the each task;

steps implemented during execution of the each task;

actors that executed the each task.

7 . The data processing platform according to claim 1 , wherein the computing server is further configured to

execute the data processing process in response to receiving a data processing request from an external agent,

reject the data processing request if, during a first predetermined time interval to a current date, a total number of data processing requests received is greater than a first predetermined threshold.

8 . The data processing platform according to claim 1 , wherein the compliance server is further configured to reject said request from the computing server if, during a second predetermined time interval to a current date, a total number of requests received from the computing server is greater than a second predetermined threshold.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 11, 2024
From: LECROART, YANNICK; AYOUBI, KAMAL; MAISONNASSE, LOÏC
To: ATOS FRANCE
Reel/Frame 067080/0759 →
Priority Claims (1)
EP 23305547 · Apr 12, 2023 · regional
Continuity (1)
Related Publication 20240348633A1 · Oct 17, 2024
References Cited (14)
US 11444966B2 · O'Hara · 2022 [cited by applicant]
US 20140351820A1 · Lee · 2014 [cited by examiner]
US 20160021138A1 · McGloin · 2016 [cited by examiner]
US 20180268491A1 · Cuomo · 2018 [cited by examiner]
US 20200067838A1 · Perumal · 2020 [cited by examiner]
US 20200074410A1 · Binder · 2020 [cited by examiner]
US 20200175195A1 · Nomura · 2020 [cited by examiner]
US 20210150411A1 · Wilhelmus et al. · 2021 [cited by applicant]
US 20210211468A1 · Griffin · 2021 [cited by examiner]
US 20210256421A1 · Sundaresan · 2021 [cited by applicant]
US 20230199004A1 · Texada · 2023 [cited by examiner]
US 20240193546A1 · Hooker · 2024 [cited by examiner]
European Search Report and Written Opinion issued in EP23305547 on Sep. 6, 2023. [cited by applicant]
Preuveneers Davy et al: “Chained Anomaly Detection Models for Federated Learning: An Intrusion Detection Case Study”, Applied Sciences, vol. 8, No. 12, Dec. 18, 2018. [cited by applicant]