- One of the largest open-source AI training data sets contains millions of images of personal documents like passports, credit cards, and birth certificates.
- This raises concerns about the privacy and security of individuals whose data may have been included in the dataset.
- The discovery highlights the need for better safeguards and ethical considerations in handling sensitive personal information for AI training purposes.
Read next
Nvidia CEO Awaits US Approval for H20 China Licenses
July 17, 2025
Nvidia's CEO, Jensen Huang, is expecting to receive the first round of US licenses to export H20 AI chips to…
xAI’s Grok 4.1 debuts with improved accuracy and new API pricing
November 20, 2025
xAI has launched Grok 4.1, featuring a reduced hallucination rate for enhanced performance on web and app…
Gemini replaces Google Assistant in Android Auto rollout
November 21, 2025
Google's Gemini is now being globally rolled out to Android Auto, offering drivers a new voice-activated…
Mastercard debuts Agent Pay, integrating OpenAI and Anthropic into its network for seamless AI-driven transactions, with partnerships from Microsoft and IBM; utilizes tokenization and AI for secure workflows and fraud prevention without window switching
April 30, 2025
Mastercard launches Agent Pay, integrating AI platforms like OpenAI and Anthropic into its payment network,…