The archive
178 stories, newest first — page 2 of 8.
AI Governance & Ethics
A small company does not need a binder of AI policies. It needs retrievable records showing who approved the system, how it was tested, what counts as an incident, and what changed.
Irene Vasko · 8 min read
AI Governance & Ethics
Selection-rate testing can reveal disparities in one hiring workflow. It cannot certify accuracy, fairness or compliance when the software, applicant pool or deployment has changed.
Irene Vasko · 8 min read
Consumer AI Hardware
Use an isolated connection and one harmless test phrase to check who controls the toy, what it records, whether mute works, and how deletion behaves.
Devin Oyelaran · 7 min read
Agentic AI & Orchestration
An ordering agent can retry after a timeout and buy the same item twice. A durable idempotency key lets the order and payment systems recognize one purchase intent across repeated calls.
Mara Quintero · 8 min read
AI Industry & Models
One AI request can have different boundaries for inference, logs, abuse review, and support. The deployment setting and access controls matter more than the API hostname.
Tobias Lund · 8 min read
AI Governance & Ethics
A disclosure that appears on the approved political ad can vanish during a routine crop or reupload. Compliance teams need records for every exported variant, not just the master.
Irene Vasko · 7 min read
AI Governance & Ethics
A disclosure in the caption or first frame can disappear while the synthetic video keeps moving. Durable labeling requires visible, audible, and machine-readable signals that fail differently.
Irene Vasko · 8 min read
AI Industry & Models
Native PDF support removes a conversion step, but it does not guarantee correct rows, footnotes, or chart labels. Use a structured parser when those relationships determine the answer.
Tobias Lund · 7 min read
AI Governance & Ethics
Colorado’s AI law turns ordinary system documentation into compliance evidence. Start with one record per decision workflow, including its inputs, reviewers and vendor dependencies.
Irene Vasko · 8 min read
Consumer AI Hardware
An NPU can run supported AI models efficiently, but the laptop badge proves little about your software. Test the exact app, model, and battery workload before paying extra.
Devin Oyelaran · 7 min read
AI Governance & Ethics
A policy statement cannot prove that an AI control operated. Auditable governance ties each model release to tests, ownership, approval, deployment records, and a defined stop condition.
Irene Vasko · 8 min read
Agentic AI & Orchestration
A successful tool call can leave the requested job unfinished. The reliable check reads the systems that own the order, warehouse task, or other business outcome.
Mara Quintero · 8 min read
AI Governance & Ethics
A support prompt retained for testing can cross from service delivery into product improvement. Approval should depend on purpose, controls, deletion coverage, and evidence.
Irene Vasko · 8 min read
AI Industry & Models
Changing the model name proves that an API call still runs. A five-gate rehearsal shows whether refusals, truncation, images, token counts, and failures still fit the application.
Tobias Lund · 8 min read
Agentic AI & Orchestration
A successful API response can carry empty, stale, or incomplete data. Test whether an agent notices the defect before grading whether it finished the task.
Mara Quintero · 8 min read
AI Governance & Ethics
A national campaign cannot treat a political deepfake disclosure as one line of boilerplate. The trigger, wording, placement, timing, and enforceability change across state lines.
Irene Vasko · 8 min read
AI Governance & Ethics
An official AI chat can leave four record candidates: the prompt, uploaded source file, generated draft, and vendor log. Agencies need retention and export controls before employees start using it.
Irene Vasko · 8 min read
AI Industry & Models
Mirror sampled production requests to a candidate model, hide its outputs, and compare each run. The method exposes task-specific regressions before a model swap reaches users.
Tobias Lund · 8 min read
AI Industry & Models
Native PDF input removes a preprocessing step, but it does not remove layout errors. The right input format depends on whether the document contains tables, columns, scans or citation-sensitive text.
Tobias Lund · 8 min read
Consumer AI Hardware
An AI toy’s parent dashboard matters more than its character voice. Check muting, conversation limits, offline behavior and deletion before the toy enters daily use.
Devin Oyelaran · 8 min read
AI Governance & Ethics
A New Jersey employer’s obligations can change when an applicant crosses a state line or encounters a different screening tool. Build the notice map around each decision point.
Irene Vasko · 8 min read
AI Governance & Ethics
A visible badge, file metadata, generation log, and signed Content Credential answer different questions. Cropping and reposting expose the gaps between them.
Irene Vasko · 8 min read
AI Governance & Ethics
There is no blanket U.S. disclosure rule. A practical answer depends on where the customer is, what the bot is doing, and whether chat becomes an AI-generated call.
Irene Vasko · 8 min read
AI Industry & Models
Moving model names remove upgrade work, but they also weaken regression evidence and incident replay. Here is when to pin a snapshot and how to stage the next one.
Tobias Lund · 8 min read