Month: August 2024

VIEW THE RECORDING

TITLE: Responsible AI in Assessment
DATE: Sep 10, 2024
TIME: 4:00–4:30 p.m. EDT

DETAILS:
AI offers significant opportunities for assessments by enabling secure, remote tests that expand access. However, it also poses risks, including the potential for new cheating behaviors, such as using generative AI tools like ChatGPT for writing assistance. As these AI capabilities continue to disrupt traditional assessment methods, integrating Responsible AI (RAI) practices into assessment development and policy becomes crucial. RAI is a global concern, and governments around the world have developed guidelines aimed at mitigating potential harms associated with AI. These guidelines are grounded in ethical principles and aim to identify risks and develop practices that uphold the quality of tests. The webinar highlights the need to develop further and expand RAI principles for assessment practices.

VIEW THE RECORDING

TITLE: Improving Language Accessibility VIA Technologically Mediated Lexical Simplification
DATE: Aug 13, 2024
TIME: 4:00–4:30 p.m. EDT

DETAILS:
During this webinar, panelists will discuss how Mediated Lexical Simplification can help to improve Language Accessibility. Lexical Simplification (LS) is the task of automatically improving the accessibility and readability of any given text by replacing hard-to-read words with simpler alternatives. LS systems achieve this by leveraging recent advances in AI, including large language models (LLMs), such as GPT 3.5 and others, to suggest several simplified alternatives to an identified complex word that maintains the original text’s meaning. Various LS systems have been designed to improve text accessibility for differing target demographics, including second-language English learners, individuals with a reading disability, or adults with low literacy. However, more collaboration is needed between AI researchers and real-world classroom environments to determine the effectiveness of LS systems.