Apple has released four audio recordings along with a research summary from its 2026 Workshop on Privacy-Preserving Machine Learning & AI. Here are the specifics.
Apple has shared a fresh entry on its machine learning blog showcasing four prominent presentations from its 2026 Workshop on Privacy-Preserving Machine Learning & AI.
During this two-day gathering, Apple researchers and attendees from the wider research community examined “the latest in privacy-preserving ML and AI,” with an emphasis on Private Learning and Statistics, Foundation Models and Privacy, and Attacks and Security.
Here’s what Apple had to say about the event:
Discussions and presentations at the workshop delved into advancements and unresolved issues in privacy and ML, covering topics like federated learning, statistical learning, trust models, attacks, privacy accounting, and the distinct challenges posed by foundation models. These research domains support innovation through thorough privacy and security assessments, linking theoretical frameworks with practical applications.
In its blog entry, Apple spotlighted four presentations, one being the ‘Crypto for DP and DP for Crypto’ talk, delivered by the company’s Research Scientist Kunal Talwar.
Moreover, other highlighted presentations include:
– Online Matrix Factorization and Online Query Release, given by Aleksandar Nikolov from the University of Toronto
– Learning from the People: Discussing S&P Technology for Responsible Data Collection, presented by Elissa Redmiles from Georgetown
– Understanding and Mitigating Memorization in Foundation Models – presented by Franziska Boenisch from CISPA
Apple also showcased 24 published works presented at the workshop, including three papers authored by current and former researchers at the company.
To view all the sessions and access the complete list of referenced papers, follow this link.