Compiled by the editorial desk with reference to official statements, public reports, and industry data.

Microsoft has removed a facial recognition dataset comprising more than 10 million images of about 100,000 individuals, most of whom are public figures, according to a report from Engadget. The dataset, known as MS Celeb, was taken down after a researcher flagged that the people in the images had not given their consent for their biometric data to be used.

The removal comes at a time when Microsoft has been vocal about the need for oversight of facial recognition technology. In November 2018, Microsoft President Brad Smith warned that the world stands at the threshold of technology that could allow governments to track individuals anywhere, and he called for regulatory measures.

MS Celeb was introduced in 2016 and was described as the largest publicly available facial recognition dataset at the time. It was intended for training algorithms used in image captioning and video analysis, as outlined in a paper by Microsoft Research.

The images were sourced from Creative Commons databases, but the subjects were not asked for permission. The Financial Times reported that a Microsoft statement said the site was meant for academic purposes and was operated by a former employee, adding that it has since been taken down.

The dataset was one of three similar collections hosted by academic institutions, including Duke and Stanford universities. The other two were also brought to light by Adam Harvey, a Berlin-based artist and researcher who investigates surveillance technology.

Harvey criticized Microsoft for using the term 'celebrity' to include people who simply have an online presence, noting that many of the individuals listed are vocal critics of facial recognition technology. He said that Microsoft was using their names and biometric information to build tools that those individuals oppose.

The removal of MS Celeb highlights the growing scrutiny of facial recognition technology and the data used to develop it. While the dataset was intended for research, the lack of consent raises questions about the ethics of using publicly available images for biometric analysis.

Microsoft has not commented further on the removal beyond the statement provided to the Financial Times. The company's earlier call for regulation suggests a shift in its stance on the technology, though it continues to develop facial recognition products for commercial use.

This incident underscores the tension between technological advancement and privacy rights, a debate that is likely to intensify as facial recognition becomes more widespread.