This article from MIT Technology Review’s newsletter, The Download, covers two major developments in artificial intelligence. First, it details how AI is being leveraged by cybercriminals to create more sophisticated and scalable attacks, including advanced phishing campaigns and automated vulnerability discovery. Second, it explores the emerging field of secure AI assistants, highlighting research and development efforts focused on building AI systems with enhanced privacy and security safeguards to prevent data leaks and misuse. The piece presents a balanced view of AI’s dual-use nature, examining both its potential for harm in the wrong hands and the technological countermeasures being developed to ensure its safe application. Read the full article for more details.