**[Important News from Last Night and This Morning (September 25 - September 26)]** **Dragonfly Partner: AI Will Not Become Safe on Its Own, Humanity Must Address Safety and Control Issues** Dragonfly Managing Partner Haseeb posted on the X platform, stating that throughout human history, we have faced numerous risks such as tiger attacks, the Black Death, smallpox, famine, earthquakes, and wars. However, these problems did not disappear on their own; they were gradually resolved through measures like vaccines, sewer construction, alliances, and earth science research. Haseeb emphasized that just because humanity has solved many risks in the past does not mean that AI risks will disappear on their own. AI will not become safe on its own; humanity must genuinely address AI safety issues and maintain control over AI. He noted that humanity may soon delegate control over almost all critical aspects of society to AI, so it is imperative to ensure that humans still retain control when that time comes. **Sources: Federal Reserve Plans to Raise Bank Regulatory Thresholds** Four informed sources revealed that the Federal Reserve is formulating a plan...