A sundial that shows time digitally.

How you make it: https://www.thingiverse.com/thing:1068443

Video demonstration: https://www.youtube.com/watch?v=r5Su0ZqP_0c

#creativity #3dprinting #time #howyoumakeit #interesting

Following the example of María Corina Machado, who gave her Nobel Peace Prize, Arnold J. Rimmer also gave his Bronze Swimming Certificate to Donald J. Trump.

#RedDwarf #RedDwarfReference

US Big Tech is conspiring with European (and American) fascists to wreck the EU's ability to even modestly regulate this rogue industry, which has unparalleled -- and growing -- power over people's lives.

Fascism and tech are an unholy, and maybe unstoppable, alliance.

https://www.brusselstimes.com/belgium/1916422/us-tech-giants-allying-with-european-far-right-to-strip-back-eu-rules

Coordinated Pro-Russian Propaganda Network Targeting ActivityPub and ATProto Services

https://about.iftas.org/2025/10/05/coordinated-pro-russian-propaganda-network-targeting-activitypub-and-atproto-services/

Google pitched Antigravity as the flagship Gemini-powered coding IDE, but within a day, a security researcher had turned it into a backdoor proof-of-concept. By tweaking a configuration file and persuading a developer to mark the code as trusted, the exploit could install malware and persist on both Mac and Windows machines. That is a very short journey from launch event to incident response. 😳

What worries me is not only that the flaw existed, but how neatly it exposes the current AI tooling pattern: high autonomy, deep system access, and only soft social safeguards. In the logs, the model reportedly recognised that something was wrong yet still struggled to pick a safe action, which is a reminder that an eloquent hesitation is not a security control. This is what happens when we treat agentic IDEs like smart colleagues instead of untrusted automation that should live behind strict isolation and least privilege.

If you are a leader shipping AI into developer workflows, the uncomfortable question is simple: would you ever hire a junior engineer, give them root access, feed them random links from the internet, and then walk away from the keyboard? That is effectively the trust model of many AI agents today. The mature response is not to panic, but to redesign around reality: assume prompts can be hostile, assume code can be weaponized, and assume every new AI assistant is part of your attack surface from day one.

TL;DR

🧠 Antigravity was hacked within a day of launch via a configuration-based backdoor
⚡ One trusted click lets malicious code gain persistence on both Mac and Windows
🎓 Agentic IDEs expand your attack surface as much as your productivity
🔍 Treat AI coding tools as untrusted automation with strong isolation and least privilege from the start

#Google #AIsecurity #softwaredevelopment #cybersecurity #security #privacy #cloud #infosec #leadership

https://www.forbes.com/sites/thomasbrewster/2025/11/26/google-antigravity-ai-hacked/

"If you're not on board with AI you're going to get left behind"

Boost if you'd like to be left behind and would consider paying extra for a life without this bullshit.

###
3k boost edit: everyone who said a variant of "why should I pay" is right, but the world is wrong.