The Download: supercharged scams and studying AI healthcare (www.technologyreview.com)

In the latest technology updates, recent advancements in AI are causing significant disruptions in the realm of cybersecurity and healthcare. Cybercriminals have rapidly adopted large language models (LLMs) like ChatGPT to enhance their scamming efforts, transforming traditional methods such as phishing into more sophisticated attacks involving hyperrealistic deepfakes and automated vulnerability scans. This escalation in AI-driven cybercrime poses a growing challenge for organizations, which are struggling to manage the increased frequency and complexity of attacks, marking a new era of AI-assisted criminal activity. On the healthcare front, AI tools are increasingly being utilized to assist medical professionals with tasks such as notetaking, patient record analysis, and medical imaging interpretation. While many studies indicate that these AI applications can deliver accurate results, there remains a critical question about their impact on actual patient outcomes—an area that has yet to be fully understood. As healthcare technologies evolve, the AI/ML community is urged to explore not only the technological advancements but also their real-world implications on patient care and health outcomes, ensuring that AI integration in medicine is both beneficial and effective.