“You Should Feed the Bots:” n.d. https://maurycyz.com/misc/the_cost_of_trash/
“You Already Have a Git Server:” n.d. https://maurycyz.com/misc/easy_git/
“Sora Might Have a ’Pervert’ Problem on Its Hands,” n.d. https://www.businessinsider.com/sora-video-openai-fetish-content-my-face-problem-2025-10
GenAI in its finest.
“The Biggest Lie in AI,” n.d. https://m.youtube.com/watch?v=0Plo-zT8W9w&t=357s
Definitely for investors, this is the biggest lie, but for us, normal folks, the whole "reasoning" is the biggest lie.
The author of this video mentioned that now OpenAI will try to respond to benchmark queries with already generated, correct results to hide that GPT5 is worse on those benchmarks.
This sounds insanely stupid and we can already see that it got worse in some areas. All of this suggests that OpenAI developers have no idea how to push it further to AGI and focus on collecting money.
“The Present and Potential Future of Progressive Image Rendering,” n.d. https://jakearchibald.com/2025/present-and-future-of-progressive-image-rendering/
“Developing a Web Server with C and FastCGI,” n.d. https://jsloop.net/2025/10/27/developing-a-web-server-with-c-and-fastcgi
The code looks extreamely simple! I was not aware that such protocol exist.
Also I wonder how fast this FastCGI is as from waht I see it starts new process for each network request.
“Largest Study of Its Kind Shows AI Assistants Misrepresent News Content 45% of the Time – Regardless of Language or Territory,” n.d. https://www.bbc.co.uk/mediacentre/2025/new-ebu-research-ai-assistants-news-content
“LANGUAGE MODELS ARE INJECTIVE AND HENCE INVERTIBLE,” n.d. https://arxiv.org/pdf/2510.15511
I think more work like this will be created in the future and it clearly show that big tech feed us with lies about AI training.