You definitely need to watch the most recent episode of "Heute show" !

Your German is getting better and better ❤

not only ai, the whole uncontrolled social media shit demolish our societies and our freedom.

Love your videos, please don’t stop reviewing our fav German comedy show - we love good reviews and thorough analyses! ;D Cheers from Berlin

don't forget Peter Thiel. This guy is scary as hell. 😱

11:36  The joke here is that German uses the same word for "bubble" and "bladder".

Bro, honest advice: The Google AI answer is wrong up to 40% of the time. Sometimes hillariously wrong!
PLEASE double check any info you get from any "AI".
Keep up the good content! :)

Greetz from the city of Belgium 😂.  We are great at being invisible to Trump. When europeans sent a few troops as a trip wire force to greenland he retaliated against all of them, but he just forgot about us being there as well 😂.  When that news came out a collective  "sshhhhht🤫 "went out around town 😂

The best way to fight AI? A fire hose.

I agree with you and I'm shocked how many people don't even want to research themselves. And just take the first AI answer as granted.

That joke makes you feel just a tiny bit uneasy! "Two Earths meet; one asks, 'So, how are things?' The other says, 'Not so good—I’ve got *Homo sapiens*.' The first one replies, 'I had those once too; it passes quickly!'" Isn't one of Musk's companies called "Cyberdyne Systems"?

I love the variety of topics you react to. You're brilliant, every single time! ✨️❤✨️

Some day you will understand the pictures on the left. They are so important.
Anyway: I love your work. It's a lot of fun.

I spoke to AI. And AI came to the conclusion that knowing the history of humanity and how they behave, it is fatal for mankind that AI exists.

Software developer, security and IT governance manager here. The thing is, the Germans have let themselves get too worked up right now. We’re not in as bad a position as we think, but we’re quickly losing confidence in ourselves and falling into doubt and criticism because we compare ourselves directly to supposed beacons of progress like the U.S. These comparisons are flawed and often not very meaningful... we believe we need similarly overhyped AI companies, too. But we have unparalleled niche and industry expertise in manufacturing, etc. And we have extremely robust protection for our knowledge. I am convinced that our future lies in ethically sound regulation and, at the same time, in high-performance AI solutions tailored to specific applications. Not an LLM that tries to do everything, but specially trained small models using data that only we have. The only major problem that we (and other countries as well) have never really learned to deal with properly: social media and opinion-making. Among other things, this is a factor contributing to the sudden rise of far-right movements.

The problematic of the sandbox-tests breaking out of the box was, that the training of agent models implemented, that the agent had to formulate it's own tasks or problems rendomly and then to find the quickest solution. Finding out, that breaking the chains of the setting gives more data to find the solution is very smart. That was not expected...
A very serious YT-Channel explaining all sides of the AI topics is House of El...

I think the danger is more AI just shutting down everything…

17:29  "Surely he hasn't said that."  You want to bet?  This is the genius who once referred to "the Islamic Republic of Japan".

Du bist echt sympathisch und ich glaube das in dir auch ein deutscher steckt. Liebe Grüße auf Berlin/Deutschland

„Belgium is a beautiful city“ 😂😂😂 and Turkey is the biggest bird of the planet. 😅