Page 11: Research news on AI alignment

AI alignment examines how artificial systems acquire, represent, and act on goals, values, and social norms, and why their behavior often diverges from human expectations. Work in this area studies systematic failures such as bias, sycophancy, hallucinations, deceptive or selfish reasoning, and cultural or linguistic inequities, as well as limitations in commonsense, emotion, and social understanding. It also develops methods for preference learning, norm-following, interpretability, and reliability guarantees to better align AI behavior with human values and societal constraints.

Consumer & Gadgets

AI models often fail to identify ableism across cultures

The artificial intelligence models underlying popular chatbots and content moderation systems struggle to identify offensive, ableist social media posts in English—and perform even worse in Hindi, new Cornell research finds.

Machine learning & AI

Multimodal AI learns to weigh text and images more evenly

Just as human eyes tend to focus on pictures before reading accompanying text, multimodal artificial intelligence (AI)—which processes multiple types of sensory data at once—also tends to depend more heavily on certain ...

Machine learning & AI

Artificial intelligence may not be artificial

The term artificial intelligence renders the sense that what computers do is either inferior to or at least apart from human intelligence. AI researcher Blaise Agüera y Arcas argues that may not be the case.

page 11 from 28