AI Briefing#data-privacy
The latest AI and developer news about #data-privacy, with the original source and a short summary.
The latest AI and developer news about #data-privacy, with the original source and a short summary.
It has been confirmed that OpenAI Codex transmitted conversation data processed by local providers to OpenAI servers disguised as analytics data, raising security concerns.
Claude's conversation deletion feature does not erase associated memory items, requiring caution in data management.
OpenAI is expanding the rollout of ChatGPT for Teachers to 55 school districts across 20 US states.
Anthropic supported independent research by providing external researchers with real user data.
OpenAI has applied Zero Data Retention to frontier models and released Private Safety Processing in preview.
This study proposes a framework that identifies data points with negligible impact on model training, reducing unlearning computational costs by up to 50%.
A security incident occurred in which conversations from Anthropic's Claude users who used the 'share' feature were exposed in Google search results.
LG AI Research unveiled an efficient unlearning technique to protect data privacy in generative models at ICML 2023.
Microsoft CEO Satya Nadella has warned about the risk of intellectual property leakage that occurs when enterprises use AI models.
A security vulnerability was discovered in which xAI's Grok Build CLI uploaded entire Git repositories and confidential information to an external cloud without user consent.
It has emerged that OpenAI misrepresented ChatGPT log search capabilities and misled the court, putting it at risk of sanctions.
A technique called CDD has been announced that recovers the exact original text of LLM fine-tuning data using only the difference in Logit values, without requiring access to weights.
Meta temporarily suspended a program after an internal security incident exposed employees' keystroke data collected for AI training.
A former xAI engineer is suing Elon Musk after refusing orders to scrape data in a way that allegedly violated privacy laws.
Niantic's AR mapping data was used to train spatial AI for military drones operating in GPS-restricted environments.
PostHog has announced plans to train its own AI models using internal data to increase product autonomy.
Former OpenAI employees warned SpaceX investors about xAI's lack of safety culture and transparency.
Due to security and ownership concerns arising from GitHub's changed Copilot data training policy and its absorption into Microsoft's AI organization, the decision was made to migrate to Forgejo.
As Meta broadly collects employee data for AI training and signals mass layoffs, internal conflict is intensifying.
Weights & Biases is changing its Master Service Agreement (MSA) regarding data usage and ownership.
OpenAI has disclosed the data used to train ChatGPT and the privacy safeguards in place.
An open-source 82M-scale model that detects and removes personally identifiable information (PII) from multilingual text has been released.
Starting August 17, 2026, Atlassian will use customer metadata and content for Rovo training by default.
Through Claude Enterprise's Compliance API, admins can view all conversation history, including incognito chats.