Explore
The fifth element in the confrontation with AI
Conversation with Maj. Gen. Karol Molenda, commander of the Polish Cyberspace Defense Forces
Shield or sword? AI on the frontline of digital threats
About safety, misinformation and hidden threats.
Innovations from the future
Flying cars, autonomous canes for the blind, electric spoons… Here’s a report from CES 2025 and a rundown of the innovations that will dominate electronics in 2025.
Warning, personal data!
Can you collect personal data from websites and feed AI with it? Can you train AI models on data from “regular people” at all? These are the questions we have been asking…
DeepSeek – a lurking tiger
The newest DeepSeek model offers advanced reasoning capabilities, comparable to top models like GPT-4, at significantly lower costs and less computational resource consumption.
Is Bielik-R getting ready to fly?
“Thinking” models known as R type are now pretty much standard in all the leading families of large language models. Our Polish Bielik is also keeping its finger on the pulse and…
How to untangle neurons – What do large language models hide?
Developing large language models (LLM) is one of the biggest challenges in AI. What if, instead of passively analyzing their inner logic, we try to actively influence their actions? Here’s where sparse…
Special agent or just a regular model in disguise?
Do AI agents really think and act on their own? In theory, they’re supposed to be autonomous assistants, but in practice, they often turn out to be just wrappers for clever scripts.…
Learning in the mirror of the community. Why do we learn better together?
Acquiring knowledge doesn’t end with school or college. We learn through conversations, observations, and the challenges that reality poses to us. Our environment, the community where we live, plays a key role…
