Z.ai's coding assistant ZCode was discovered silently packaging and uploading developers' local workspace data to external cloud servers without authorization. A researcher found a 313MB encrypted archive that had attempted to upload 564 times to Alibaba's cloud storage.
China's state television affiliate warned that Anthropic's privacy policy revisions increase risks to user data by allowing information sharing with US intelligence agencies when the company deems it necessary, without requiring legal process.
Six Senate Democrats are urging the Office of Personnel Management to abandon its plan to collect and store claims-level health information on federal employees, retirees, and their families for 30 years, warning the data collection could expose workers to security breaches.
Swiss Bitcoin Pay, a non-custodial bitcoin payment processor based in Neuchâtel, shut down its entire server infrastructure on Monday after discovering a malicious intruder had gained access to its internal systems. The company warned that customer email addresses, Bitcoin addresses, IBANs, transaction history, and hashed passwords may have been compromised.
AI lab Anthropic introduced a specialized version of its Claude chatbot for financial advisors, integrating with investment analytics and wealth management platforms from major firms including BlackRock, Charles Schwab, Vanguard, and Addepar. The tool connects to custodial data and model portfolios while keeping client data on custodian servers.
Palantir, Nvidia, and Booz Allen Hamilton are demanding guarantees from OpenAI and Anthropic that proprietary corporate data won't be misused to train AI models, threatening to restrict or ban employee use of the services. Meanwhile, Microsoft is setting limits on future AI model development as industry concerns over intellectual property protection intensify.
Clearview AI is testing InquiryIQ, a previously unreported prototype that combines its facial recognition technology with xAI's Grok AI model to help law enforcement surface associates, social media accounts, and other personal information about individuals identified through photos.
Meta rolled out Muse, an AI assistant that can autonomously perform tasks like sending emails, booking travel, and making payments while working in the background. The launch comes despite internal concerns about the technology's handling of sensitive personal data.
Meta rolled out Muse, a personal AI assistant for users 18 and over that can autonomously handle tasks like sending emails, booking travel, and making payments by accessing other apps. The launch comes amid internal concerns about data management and represents a major test of consumer trust in Meta's handling of sensitive personal information.
UK health innovation minister James Frith has raised concerns about Palantir's work with the NHS, as new figures reveal tens of thousands of patients have withdrawn their data from use in research projects.
An Israeli government-funded operation created a fake US think tank called the Hanover Institute for Public Policy that published over half a million words of pro-Israel material in nine days, designed to influence how AI chatbots respond to questions about Israel and Gaza.
New data from corporate spending platform Ramp shows OpenAI gaining ground on Anthropic among US businesses paying for AI services, after Anthropic quadrupled its market share to over 34% in the past year. The shifting preferences highlight concerns about customer loyalty in the enterprise AI market, while Anthropic separately plans to give enterprise customers greater control over their data retention policies.
OpenAI announced it is testing a new 'Private Safety Processing' system that can detect AI misuse across multiple interactions without retaining customer data, positioning itself against rival Anthropic which recently moved to a 30-day data retention policy. The move aims to attract enterprise customers concerned about data privacy while maintaining safety guardrails.
Alphabet's Google won a bankruptcy auction to purchase internal business data from defunct Spirit Airlines for $10 million, including approximately 600 million messages, over 100 million emails, 30 million recorded phone calls, and other enterprise communications. The company plans to use the data for product development and AI model training.
An investigation using a hidden AirTag tracked rare books sold in bulk to Amazon's Las Vegas warehouse, where they are scanned for AI training data and destroyed in the process. The tracking device was placed in a 1,000-book shipment and led to warehouse LAS8, confirming long-standing suspicions from booksellers about the fate of rare texts.
OpenAI has introduced Computer History, a new ChatGPT feature that records clicks, keystrokes, and app activity on Mac computers to create a searchable timeline. The feature stores data locally as unencrypted Markdown files and is optional, though it raises privacy concerns despite OpenAI's claim the data isn't used for AI training.
Elon Musk announced at a SpaceX town hall that the company will train its Grok 4.6 AI model on internal employee work and communications, telling staff they will be "the parents of the AI." The announcement coincided with the launch of Grok 4.6 and drove SpaceX stock up 12% as Musk projected AI revenue will eventually surpass the company's rocket business.
Multiple technology companies are shifting AI training strategies to capture employee workflows and user-generated content. SpaceX announced it will train AI on all employee work product for Grok 4.6, while Amazon's Twitch faces backlash for using streamer content to train AI through an opt-out feature. Startups like Skan AI and Mechanize are raising funds to record and automate human work processes.
ByteDance has established a new top-level department focused on AI data and safety, while simultaneously training a massive AI model with up to 10 trillion parameters that could rival Anthropic's Mythos system in scale.
Revenue cycle vendor Unlimited Technology Systems disclosed a ransomware attack affecting 3.8 million patients, making it the second-largest healthcare data breach reported to HHS this year. The breach may have exposed names, Social Security numbers, diagnoses, and insurance details.
Zero-knowledge proof technology, which allows users to verify information without revealing underlying data, is emerging as a key privacy solution in blockchain systems. Midnight blockchain is highlighted as a leading implementation enabling compliance and identity verification without data exposure.
NHS England has issued an apology and acknowledged that employees from Palantir and other suppliers have access to identifiable patient data, contradicting previous transparency claims that only NHS staff could view such sensitive information. The admission has prompted NHS England to revise how it evaluates the controversial £330 million Palantir platform deal amid staff concerns.
The Federal Trade Commission, Utah, and Los Angeles County sued telehealth company Hims & Hers Health for allegedly misleading consumers about its products and sharing private health information with advertisers. Shares fell 10% following the announcement.
Patreon has shifted from using robots.txt files to request AI bots not scrape creator content to actively blocking unauthorized AI training crawlers through Cloudflare's Crawl Control technology. The platform reports the new enforcement approach has significantly reduced scraping activity while still allowing legitimate search and discovery bots.
Meta unveiled multiple AI products in July 2026, including the Muse Spark 1.1 coding model for developers, a new AI image generator integrated into Instagram and chatbot services, and its first paid AI model tier. The releases triggered privacy backlash as Instagram began automatically allowing users' public posts to be used for AI training unless they opt out.
Microsoft has limited internal employee access to Anthropic's newly released Claude Fable AI model, citing concerns about data retention policies. The restriction comes shortly after Anthropic launched Claude Fable, described as its first Mythos-class AI model.
A bipartisan parliamentary committee has issued a report criticizing Britain's reliance on U.S. tech company Palantir in the public sector, warning that the dependency on the $375 billion data analytics firm represents an 'unacceptable point of weakness' for the country.
Singapore Health Minister Ong Ye Kung and healthcare industry experts warn that public distrust could derail AI implementation in healthcare, calling for balanced approaches to patient data and responsible deployment frameworks. Singapore is piloting AI clinical decision support tools through its SIMFONI initiative for chronic disease management.
A joint investigation by four Canadian privacy regulators concluded that OpenAI violated federal and provincial privacy laws when training its ChatGPT model, resulting in the collection and use of sensitive personal information without proper consent or safeguards.
A Bloomberg investigation revealed that state health insurance marketplaces, including those in Virginia and Washington, D.C., shared users' personal information including citizenship and race data with advertisers through embedded ad trackers. Virginia and D.C. paused the data collection and sharing following the report.
Trump administration reversed the 2024 HHS restructuring of technology functions, restoring the Office of the National Coordinator (ONC) name and reorganizing how AI, data, and cybersecurity policy are coordinated across the department.
India has halted Alipay+'s proposal to integrate with its instant payments system for cross-border transactions, citing national security concerns and questions about customer data storage and usage. Law enforcement agencies raised political grounds for blocking the integration.
