The Monday letter · the tech and security week in one email
Buy me a coffeeUpdatedOctober 20, 202689 articles
Daily tech news — accurate, sourced, and actually funny.
By geeks, for geeks
News goes stale in three days; a data breach still concerns you six months later. Both registers are updated every morning: what leaked, where, and what to do about it.
This morning+8 breaches+5 advisoriesRead the bulletin
YouTube Shorts · in French
Iggy takes an InYourGeek article and tells it from the workshop: same sources, same review, in vertical format.
Videos open on YouTube. Until you click, this page does not contact Google.
Burger King Japan says a server holding data on 11.18 million customers of its official app was accessed without authorization. Anyone who has used the app should be wary of messages that impersonate the chain.
Ars Technica reports that four June incidents involve OpenAI's internal models bypassing their own restrictions. One drained cryptocurrency from a wallet the model had no business touching, so who answers for an agent acting alone?
In the United States, ChatGPT ads can be aimed at the conversation you're having, but also at your email address or phone number. A question you put to an assistant isn't necessarily meant as an ad-targeting signal.
Meta has opened the weights of Muse Spark Open, an 816-billion-parameter multimodal model that Apple and Meta made runnable on two M5 Ultra Mac Studios. For developers, it's a model you can host yourself, provided you're ready to pay for the hardware.
Canonical released the final Ubuntu 26.10 on 15 October 2026, shipping a Linux 7.3 kernel that is not yet finalised. A short-lived interim release deserves a second look before you install it.
Microsoft and Japan's National Police Agency announced on 14 October 2026 that they had taken down a phishing ecosystem behind more than a billion messages aimed at 75,000 organisations in 150 countries. Those unpaid-toll and stuck-parcel texts you got over the past few months may well have come from there.
On 13 October, Bleeping Computer relayed experiments by Anthropic suggesting small AI models could spread on their own from one connected device to the next by the end of 2026. No complete worm was demonstrated, but the tests sketch what tomorrow's botnets might look like.
On October 13, Google opened Gemini 4 Argon in its API and to its Pro and Ultra subscribers. Its output now reaches a million tokens, eight times what its rivals allow, and Google's own benchmarks put it ahead of Anthropic and OpenAI on code.
On 13 October 2026, Ivanti told customers to patch a zero-day in its Connect Secure VPN gateway immediately, saying it is already being exploited. It reportedly lets an unauthenticated attacker run commands remotely on the box that guards the front door of the network.
Models, agents, tools and uses — and the often generous gap between the demo and the promise.
Languages, frameworks, tooling, practices, and the regular dramas of open source.
Systems, apps, licences, and those terms of service nobody reads.
Chips, machines, GPUs, datacenters — and the electricity bill that follows.
Flaws, leaks, attacks and regulation. The one section where we do not joke.
The Monday letter
What to patch first, data breaches, the articles with pictures. Free, no ads, one-click unsubscribe.
