As AI adoption continues to grow, propelled by the frenzy around Generative AI, the need for human involvement and oversight into the development of the technology has never been greater to ensure that it is safe, reliable, and accurate. While evaluating AI data pipelines for bias and diversity gaps has been a best practice for years, it is increasingly critical for organizations to invest in the right training data to optimize their solutions. Having a clear understanding of how AI data is collected, processed and annotated leads to greater trust in AI model results, which in turn builds trust with end users.
LXT AI data experts Julia Liberty and Jessica Fernando have each spent many years guiding companies through the process of designing data collection and annotation programs that result in high-quality training data for a wide variety of AI solutions. They recently discussed the role of AI data strategies in developing and enhancing AI products in this LinkedIn Live webinar.
In this webinar you will learn:
- The critical role of humans in developing AI training data
- The role of a data partner in the AI product development and enhancement life cycle
- Details about a range of data projects including:
- Utterance collection
- Child speech data collection
- Toxic language detection
- How gender differences impact data evaluation tasks
Check out the 30 minute recording here to learn from our experts.