NEWS

Zambrano’s talk highlighted how financial institutions depend on machine learning and statistical models for high-stakes decisions, why model failures can lead to real financial and regulatory harm, and how strong model risk management is essential for safe, responsible innovation in banking.

We are proud to share that Dr. Depeng Xu’s paper, “Fine-tuning LLMs with Cross-Attention-based Weight Decay for Bias Mitigation,” has been accepted at EMNLP 2025, one of the top-tier international conferences in natural language processing.

On July 7–8, 2025, the Center for TAIMing AI at UNC Charlotte hosted a two-day workshop titled “Mastering Modern AI” at the Dubois Center in Uptown Charlotte. Led by Dr. Agus Sudjianto, Executive in Residence at the UNC Charlotte School of Data Science, the workshop focused on advanced techniques for machine learning (ML) model development […]

In this seminar, Schmidt explored how AI deployed without consideration of how to measure, manage, and mitigate risks can do real harm, especially in sectors such as finance, healthcare or insurance.

In this seminar, Matt Bedsole explored how Lowe’s has approached developing a scalable model monitoring framework to maintain AI reliability in production. He discussed key challenges in AI observability—from detecting data drift to improving model explainability—and shared real-world insights on operationalizing monitoring at scale.

In this seminar, Dr. Miroshnikov explored algorithmic approaches to fair lending that promote more transparent and accountable financial decision-making.

Several members of TAIMing AI, alongside leading industry researchers, gathered at the Joint Mathematics Meetings 2025 in Seattle for a special session on Trustworthy AI Applications.

In this seminar, Dr. Gurcan discussed the transformative impact of AI in healthcare, highlighting its potential to reduce diagnostic disparities, enhance treatment strategies, and address challenges like workforce shortages and health inequities

The Center for TAIMing AI published its inaugural newsletter on November 26, 2024, the newsletter included the following note from the Director

Prof. Jake Lee delivered a seminar that explored the delicate balance between privacy preservation and trustworthiness in example-based learning, highlighting the challenges and opportunities this approach offers for AI systems. He discussed how example-based learning enhances AI explainability and efficiency, while also addressing concerns about data privacy and the need for robust safeguards.