Finger Lakes Community College holds inaugural webinar on AI and bias

Deborah Ortloff, co‑leader of the AI initiatives and FLX AI Hub at Finger Lakes Community College, opened the webinar to introduce the series and the session topic: bias in artificial intelligence. "My name is Deborah Ortloff," she said in the event opening, and she turned the session over to computing science professor Dave Gadeau for a detailed primer.

Gadeau framed AI systems as statistical, data‑driven tools and summarized training at scale: models ingest vast collections of text, images and audio and learn probabilistic associations. "So, AI is nothing more than a probability machine," he said, adding that sampling among high‑probability continuations explains variable outputs. The webinar moved from basic mechanics to concrete examples—image and language biases, medical label confusion, hiring algorithms, platform cropping, and geographic limits for autonomous vehicles—before closing with mitigation recommendations and a Q&A.

The session recommended practical steps for users and organizations: review and sanitize sources, use inclusive LLMs (Gadeau cited a startup called Latimer as an example), run controllable local models when privacy or domain specificity matters (NotebookLM was mentioned), and use feedback controls like thumbs up/down to signal problematic outputs. Ortloff closed by inviting participants to FLCC's ai resources and the next monthly webinar.