Johnny Bang

Large language models (LLMs) showcase significant capabilities in reasoning and automation. However, the transition from an impressive demonstration to a functional product relies heavily on how effectively these systems learn from real user interactions. The integration of feedback loops is a crucial aspect often lacking in most AI deployments. As LLMs find their place in […]

Researchers have presented a comprehensive survey of “OS Agents,” a category of artificial intelligence systems designed to autonomously control devices such as computers and mobile phones. The study, spanning 30 pages and accepted for presentation at the Association for Computational Linguistics conference, outlines the rapid development of these systems, which have garnered significant investment from

Equity trading, while heavily regulated, has often led the way in technological innovation within the financial services sector. However, many banks are exercising caution in the adoption of artificial intelligence (AI) applications and agents. A recent example is TD Securities, the trading division of TD Bank, which introduced its TD AI Virtual Assistant on July

Researchers at Salesforce and the University of Southern California have introduced a novel technique enabling computer agents to execute code while interacting with graphical user interfaces (GUIs). This system, termed CoAct-1, integrates coding capabilities with cursor manipulation and button clicks, aiming to enhance workflow efficiency and minimize errors. CoAct-1 aims to overcome the limitations of