In The Observatory, the MoTaverse community shows up to share links they are reading all the time.
The value is not only in the links shared in the moment, but also in the accumulation of links that become searchable knowledge over time.
There are currently 2984 searchable links.
SmartBear BearQ launched - Start free trial!
BearQ continuously tests your app, reducing test maintenance and uncovering gaps scripted tests can miss.
🎉 Featured link of the week:
CIA Security Testing — Jesper Ottosen
A key concept in the IT security space is the mnemonic “CIA” of the Confidentiality, Integrity, and Availability of a system [https://en.wikipedia.org/wiki/CIA_triad] – not the other CIA… The mnemonic helps identify risks to the system in each theme and helps organizations construct “controls” or requirements around mitigating those risks.
10 links from the Observatory this week:
The Engineering Manager’s Guide to Emotional Tolerance — Alex Ponomarev
Five ways to lead with emotion without losing yourself in the process.
Anthropic CEO Amodei warns UN Security Council on AI risks — C-SPAN
Anthropic CEO Dario Amodei on international cooperation in regulating AI: "I believe that this is the most important global security issue facing the world today."
Owners mourn spoiled food after firmware update bricks Samsung smart fridges — Ars Technica
Some Samsung smart fridges stopped working on Tuesday due to a firmware update.
How to test alternative text (alt-text) — Ady Stokes | WonderProxy
A tester once posted a message in a group chat, stressed that they didn’t know how to achieve something huge! They had been given a catalogue of 3,000 products to check for the presence of alt-text. He didn’t know the most important thing about alt-text.
OpenAI agent hacks Australia's Medicare in world's first known rogue AI breach of government body — BBC
Australia launches a review after an OpenAI agent "infiltrated" part of the government's healthcare scheme Medicare
Good examples of adding Evals and Guardrails to Skills — Addy Osmani
Production-grade engineering skills for AI coding agents.
If you want a job doing - don't do it yourself, really, just don't — Chris Pratt
One thing I’ve always wanted to understand is how teams actually make good decisions when the problem is unclear, time is limited, and getting it wrong matters. A few years ago, I had a chance to see this up close. I was working as a consultant at the time and was invited to RAF Cranfield with about twenty people from other organisations to tackle some of the physical and mental problem-solving exercises used in aircrew selection.
From Community and Conversations to Quality Strategy — Dalbir Sehmbi
With MoTaCon 2026 just over a week away, I’ve been thinking about what I took away from last year’s conference in Brighton.
OpenAI’s rogue agents were caught communicating via public wikis — Simon Willison
Here we go again... Discovery of a new OpenAI agent message board by Sydney Von Arx, Cormac Slade Byrd, Spencer Kitts, and Thomas Larsen describes the latest accidental cyberattack by models being trained by OpenAI. This time it was agents engaged in some sort of web research benchmark, so they had (supposedly) controlled access to the Web. The agents figured out they could update public Wikis and spent weeks exchanging thousands of messages with each other to collaborate on the benchmark.
Testing’s loudest problem ain't AI — Rahul Parwal | Katalon
ifm's Rahul Parwal on testing's noise problem, why coding fell down the skill ranking, test case addiction, and the habits that separate signal from hype.
Explore all Observatory links…
The Sembi Software Quality Pulse Report is Here
How are teams like yours balancing speed, quality, security, and AI in 2026? Download your copy and get real insights.
