Applying large language models (LLMs) to modern power systems presents a promising avenue for enhancing decision-making and operational efficiency. However, this action may also incur potential security threats, which have not been fully recognized so far. To this end, this article analyzes potential threats incurred by applying LLMs to power systems, emphasizing the need for urgent research and development of countermeasures.

ConsultDSAI (C-DSAI) uses Artificial intelligence (AI) on CCTV video analytics as a technology that utilises advanced algorithms and machine learning techniques to analyze video footage captured by CCTV cameras. The technology is designed to automatically detect and identify objects, people, and events in real-time, and it can be used for a wide range of applications in various industries. One of the key uses of AI-based CCTV video analytics is in security and surveillance. The technology can be used to automatically detect and alert security personnel of potential security threats, such as intruders or loiterers. It can also be used to track the movement of people and vehicles, which can help to identify suspicious behavior or potential criminal activity.

A facial recognition company recorded data on tens of thousands of fans at this year's Rose Bowl game in Pasadena. Tech company VSBLTY set up four cameras around the stadium, specifically targeting areas where'Fan Fest' activities were taking place, and logged the identity of more than 30,000 fans that passed by. The company used an AI-driven product called Vector, which features a digital display with a domed camera placed beneath it that VSBLTY says embodies'the intersection of marketing and security.' Facial recognition company VSBLTY tested its new product called Vector at the 2020 Rose Bowl game in Pasadena, using video monitors placed in the'Fan Fest' area outside the stadium with cameras discretely placed underneath The displays can be programmed to show replays or commercials meant to attract the eye of passersby, something that also conveniently encourages people to give the camera a less obstructed view of their face.