CIRCLE WITH A DOT

@AAKL I'm taking inspiration from one of our greatest humorists (Chudomir), who wrote some of his best jokes ("Scribbles in the hospital") while dying of cancer.

@bontchev alt-text: tweet from user klara_sjo that reads "There will be no WW3. They've abandoned numbered releases and switched to a live service model with seasonal events."

@tuxicoman @bontchev yeh, but I'm not even going to read what you wrote, because Substack platform people who want to kill me and my people.I don't platform Nazis, even vicariously

"Officer reportedly leaks location of French aircraft carrier with Strava run":https://www.bbc.com/news/articles/cd9vdel17wqo

The situation with the war on Iran.

@AAKL Dunno; I haven't worked in the industry for 2 decades already. I'm in the academia, about to retire.The point I was trying to make is that, at least for programming tasks, the models are *very* different. ChatGPT wasted me a whole day before it reached the conclusion that what I wanted couldn't be done. (It kept insisting that it can be.) Claude is very good. Despite the occasional mistake or misunderstanding, talking to it feels like talking to a person. It really helped me produce useful stuff. ChatGPT feels just like chatty computer and produced mostly bullshit.

"Instagram to discontinue end-to-end encryption for DMs":https://www.androidpolice.com/instagram-is-getting-rid-of-end-to-end-encryption-for-dms/

@bontchev Already crossed that bridge. And there was an article I saw yesterday that mentions a horrible potential scenario - horrible economically, environmentally, and geopolitically. https://www.theguardian.com/world/2026/mar/11/why-irans-vital-kharg-island-oil-hub-is-still-untouched-by-us-israel-bombers

@AAKL LOL, LLMs are anything but "super". They aren't even intelligent. In fact, I suspect that even if we manage to construct real (self-aware, reasoning) AI one day, it won't be an LLM.Still, they are a useful tool; more useful than googling. Whether their usefulness is worth the cost is an entirely different matter.

The universal truths of software development.

@bontchev 🤨

We made a mistake...

@bontchev

Turnabout is fair play, exhibit #3 "Iran-linked hackers target IP cameras across Israel and Gulf states for military intelligence":https://securityaffairs.com/189069/cyber-warfare-2/iran-linked-hackers-target-ip-cameras-across-israel-and-gulf-states-for-military-intelligence.html

Turnabout is fair play, exhibit #2 "Hackers Spread Fake Red Alert Rocket Alert App to Spy on Israeli Users":https://hackread.com/hackers-fake-red-alert-rocket-alert-app-spy-israel-users/

You know what would be a great propaganda move for China? If they offered free access to their satellite imagery that shows the destruction caused by Iran - imagery, that Israel, the USA, and the Arab countries are currently censoring.

The US: "Our military (that cannot protect our own radar installations from Iran's strikes) is offering protection to the ships passing through the straight of Hormuz".The oil and gas transport industry: "Thanks, but no thanks".

I hear the US military lost 3 jets to friendly fire over Kuwait. I also hear the US military is using AI. I can just see the interaction:

@bontchev Slot Machine vs. Slop Machine

"Large-scale online deanonymization with LLMs":https://arxiv.org/abs/2602.16800v1