Zoom says its "federated AI" model, combining its SLM with open- and closed-source models, got 48.1% on Humanity's Last Exam vs. 45.8% for Gemini 3 Pro w/ tools (Xuedong Huang/Zoom)

A new online exam proctoring system named iExam has been developed, which integrates lightweight face detection and deep recognition technologies to enhance the integrity of online assessments conducted via platforms like Zoom. This system addresses the challenges of monitoring multiple student video feeds in real time and includes features for post-exam analysis to detect abnormal behaviors such as face disappearance and identity substitution.

Google has launched its most advanced AI research agent to date, based on the Gemini 3 Pro model, allowing developers to integrate this tool into their applications. This release coincided with OpenAI's introduction of GPT-5.2, marking a significant moment in the competitive AI landscape.

Google DeepMind has introduced a new benchmark called FACTS, designed to assess the reliability of AI models more thoroughly. The results indicate that even leading models like Gemini 3 Pro and GPT-5.1 exhibit significant shortcomings in factual accuracy, highlighting the challenges these technologies face in delivering truthful information.

Google has introduced a new benchmark called 'FACTS' aimed at measuring the factual accuracy of generative AI models, addressing a critical gap in existing benchmarks that focus primarily on task completion rather than the truthfulness of the information generated. This initiative is particularly significant for industries where accuracy is essential, such as legal, finance, and medical sectors.