Agent's Take: George Pickens' breakout season with Cowboys puts Jerry Jones on the clock for another mega deal
https://www.cbssports.com/nfl/news…
Does somebody know this overlay thingy which annoys me since a couple of months and I don't know where to disable it.
It's not app-specific, it sometimes appears when I want to interact with something at the top of any app. It vanishes after touching somewhere else and is reduced to a white bar at the top.
I could not find out about its purpose or functionality. The middle part icons do nothing.
My environment is
El Salvador: Weltweit erstes landesweites KI-Bildungsprogramm zusammen mit xAI
Nayib Bukele und Elon Musk vereinbaren den Einsatz des KI-Chatbots Grok in El Salvadors öffentlichen Schulen. Andernorts ist Grok durchaus umstritten.
»El Salvador — Weltweit erstes landesweites KI-Bildungsprogramm zusammen mit xAI:
Nayib Bukele und Elon Musk vereinbaren den Einsatz des KI-Chatbots Grok in El Salvadors öffentlichen Schulen. Andernorts ist Grok durchaus umstritten.«
Ach du meine Güte. Wenn dies nicht Musk's Propaganda dann den Kindern durch die KI Hirnwäsche durchgeführt wird, damit sie ihm unhinterfragt folgen?!
😬
Series A, Episode 11 - Bounty
BLAKE: Keep your head, Vila! That way I might have a chance at keeping mine.
VILA: Yes. Avon?
AVON: What?
VILA: Shut up. Please.
[On the flight deck.]
SARKOFF: Blake is finished, I'm afraid.
https://blake.torpidity.net/m/111/405 …
House Dems fail to stop internal revolt on condemning Chuy García (Andrew Solender/Axios)
https://www.axios.com/2025/11/18/chuy-garcia-marie-gluesenkamp-perez-vote-election
http://www.memeorandum.com/251117/p159#a251117p159
Tim Sweeney says Fortnite will not return to iOS in Japan in 2025, as Apple is "charging a competition-crushing 21% junk fee on third-party in-app payments" (Tim Sweeney/@timsweeneyepic)
https://x.com/timsweeneyepic/status/2001494517996732598
A small number of samples can poison LLMs of any size:
https://www.anthropic.com/research/small-samples-poison
"In a joint study with the UK AI Security Institute and the Alan Turing Institute, we found that as few as 250 malicious documents can produce a "
Anthropic open sources a method to score AI model political evenhandedness; Gemini 2.5 Pro got 97%, Grok 4 96%, Claude Opus 4.1 95%, GPT-5 89%, and Llama 4 66% (Ina Fried/Axios)
https://www.axios.com/2025/11/13/anthropic-bot-bias-data