Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the critical challenges of trusting machine learning models, particularly large language models, in this conference talk from NDC Melbourne 2025. Discover how seemingly accurate ML algorithms can deceive users and learn why assessing model performance has become increasingly complex in the era of "black box" models. Examine the various ways models can produce misleading outputs and investigate practical approaches to cut through confusion and uncover the truth about model reliability. Gain insights into the essential skills needed to evaluate LLM performance accurately as these powerful but opaque systems become more prevalent in software development and business applications.
