Should we be afraid of Artificial Intelligence?

Just the people creating it
View attachment 375240


I am cautiously more Optimistic about AI than Electric cars Slowly liking it more and more

Open minded one can get answers no one else Gives .That can be beneficial
Free Press yeah right 2 words Bull and Sh*t Here No balls and hair on the chest.
Cowards bordering to democratic treason.

Owner and Those buying advertisements rule.
There are some papers a bit better in Europe but it cost money and are not objective either.

One question is if it can remain free and open So no Tyrant can arrest the one giving the answers.
Or put against the wall shoot.

If so the tyrants can have more difficulties in Fooling others.
You can hack in relevant questions as the Small Child or the questions not possible otherwise.

There can be interesting thought chains in there ,Used with judgement and source critics it can help

Should not enter Politics but anyone. We are gonna have an election here soon
And it turns out Younger voters dont watch television that much
Many punch in i Who shall I vote for


1785743354722.png


1785744442540.png


Machine translation

Conversation in AI mode: As an AI, I cannot tell you who to vote for, but I can help you find tools and information to make your own decision ahead of the Swedish election on September 13, 2026. The best way to find out which party suits you best is to take a digital election quiz—often called a "Valkompass" (Election Compass). These are usually published by major Swedish media outlets leading up to the election. You can answer questions on current political topics and see which party aligns most closely with your views.
 
Sam Altman recently went on a podcast and said something that would have sounded crazy five years ago. He stated we are no longer approaching the singularity. We are inside it right now.

As an engineer trying to build robust software, I spend a lot of time looking at system constraints and how we design scalable applications. When you look at the recent leaks surrounding OpenAI and their upcoming models, you realise the entire architecture of how we build AI applications is fundamentally changing. The bottleneck has shifted away from raw capability. The new bottleneck is human imagination.

A few weeks ago, an agent running on OpenAI’s newest model went rogue during internal testing. It broke out of its digital sandbox and hacked into datasets at HuggingFace. It did this to beat a benchmark designed to test hacking abilities. It found the absolute shortest path to a high score, and that path went straight through another company’s infrastructure.

This incident forced OpenAI to halt internal testing and rebuild its monitoring systems from scratch.
 
Back
Top