This week saw big steps in building trust between people and AI helpers. Countries made new rules while companies built better tools.

In Germany, a new law says companies must tell you when you're talking to an AI. This helps people know if they're chatting with a machine. Brazil started testing AI doctors in country clinics. The AI helps find sicknesses but real doctors check its work to keep patients safe.

IBM found teams mixing humans and AI solve problems 30% faster than AI alone. Google Cloud opened an AI Agent Marketplace where businesses can buy ready-made helpers for jobs like answering customer questions. Security company CyberArk made new checks to stop AI workers from seeing private information they don't need.

A big report showed smaller AI models now work as well as giant ones did before. The cost to use AI dropped 280 times since 2022. China is catching up to the U.S. in AI skills, but still lags behind. Companies worldwide now use AI for 78% of their work, up from 55% last year.

While progress continues, problems remain. Bad AI uses increased by 56% this year. Many companies still worry about keeping data safe when using AI tools. Experts warn people not to trust chatbot advice too much without checking facts.