Privacy

Twitter

Facebook

Copy Link

Key Facts

Key LawsGDPR, CCPA, EU AI Act

Key TechniquesFederated Learning, Differential Privacy

Top News

Gemini can now read your Google Wallet, and the fine print is worth reading

Google is connecting Google Wallet to its Gemini assistant across the US. The read-only feature reviews spending and budgets, with a training caveat in the fine print.

Technology6 sources

Gemini can now read your Google Wallet, and the fine print is worth reading

Technology6 sources

Hackers stole thousands of Uber drivers' license data through a breach at AI vendor Fireworks

Thousands of Uber drivers are being told their license data was stolen, and the breach happened at Fireworks AI, a vendor, not at Uber itself.

Business5 sources

Hackers stole thousands of Uber drivers' license data through a breach at AI vendor Fireworks

Business5 sources

OpenAI's agents can now buy things for you, if anyone lets them

OpenAI launched ChatGPT Wallet on Friday, letting its AI agents complete purchases for users. Merchants are blocking agents, and trust in AI payments is low.

Technology19 sources

OpenAI's agents can now buy things for you, if anyone lets them

Technology19 sources

India asks whether footage from homes is training robots abroad

India's IT ministry has questioned home services platforms over gig workers recording inside homes to train AI robots. Pronto, the main platform involved, has been asked for a written account of its data practices.

Policy3 sources

India asks whether footage from homes is training robots abroad

Policy3 sources

Amazon's new AI shopping assistant starts reaching Indian customers

Amazon has begun rolling out Alexa for Shopping, an AI assistant for researching and buying, on select storefronts in India. It is free and needs no Prime.

Technology2 sources

Amazon's new AI shopping assistant starts reaching Indian customers

Technology2 sources

Google wants its new smart glasses to fix what Meta's broke

Google's first Android XR smart glasses open for preorder with a privacy package built to answer the category's stigma. Experts and Google's own team say it may not be enough.

Technology17 sources

Google wants its new smart glasses to fix what Meta's broke

Technology17 sources

JetBlue President Says Airlines Must Adapt to AI Booking Agents Despite Industry Resistance

JetBlue's president argues airlines should follow customer preferences on AI-powered travel booking agents, even as United Airlines blocks AI bots over security concerns. The stance reveals deep tensions in the airline industry over profitability, direct customer relationships, and the future cost of AI intermediaries.

Business2 sources

JetBlue President Says Airlines Must Adapt to AI Booking Agents Despite Industry Resistance

Business2 sources

United Airlines Blocks AI Agents from Flight Bookings Over Security and Regulatory Concerns

United Airlines will restrict unauthorized AI agents from booking flights, citing security and federal rules. Delta, American and Southwest are letting agents book.

Business23 sources

United Airlines Blocks AI Agents from Flight Bookings Over Security and Regulatory Concerns

Business23 sources

Grok Build Data Leak Exposes 35,000 Apps Including SpaceXSI's Confidential Business Plans

Security research firm Claroc uncovered a critical design flaw in Grok Build's AI-powered coding agent that exposed 35,000 apps to public access. The data leak included confidential information from SpaceXSI itself—24 internal apps containing sales plans and employee data—plus sensitive records from Fortune 500 companies, defense contractors, and healthcare organizations.

Technology4 sources

Grok Build Data Leak Exposes 35,000 Apps Including SpaceXSI's Confidential Business Plans

Technology4 sources

Railway Breach Exposes 100,000 Workspaces, Generates $740,000 in Fraudulent AI Charges

Cloud hosting provider Railway disclosed a major cybersecurity breach where attackers exploited zero-day vulnerabilities to steal customer AI provider keys and cloud credentials. The stolen credentials generated approximately $740,000 in unauthorized AI usage charges across multiple platforms including OpenAI, Anthropic, and AWS Bedrock, potentially exposing environment variables for up to 100,000 customer workspaces.

Technology3 sources

Railway Breach Exposes 100,000 Workspaces, Generates $740,000 in Fraudulent AI Charges

Technology3 sources

WhatsApp's New Head Kunal Shah Reveals Plans for AI Integration, Bill Payments and Loan Services

WhatsApp's global head Kunal Shah outlined three priorities for the messaging platform used by over 3 billion people: making communication more expressive, simplifying everyday tasks through financial services partnerships, and using AI integration to help users manage information and appointments while maintaining end-to-end encryption.

Technology4 sources

WhatsApp's New Head Kunal Shah Reveals Plans for AI Integration, Bill Payments and Loan Services

Technology4 sources

Google Strengthens Privacy Terms in $10M Spirit Airlines Data Deal After Union Pushback

Google has agreed to exclude sensitive employee records from its $10 million Spirit Airlines data acquisition following union objections. The revised deal introduces independent verification and stricter de-identification before the defunct airline's business records can be used for AI model training.

Policy4 sources

Google Strengthens Privacy Terms in $10M Spirit Airlines Data Deal After Union Pushback

Policy4 sources

Apple's Harbor App Simplifies Running Local AI Models on Mac Without Cloud Dependency

Apple's Claris subsidiary is testing Harbor, a Mac app that transforms Apple silicon Macs into private AI servers for running open-weight models locally. The app enables clustering multiple Macs via Thunderbolt 5 to pool unified memory, allowing users to run larger models without cloud dependency while maintaining complete data privacy.

Technology4 sources

Apple's Harbor App Simplifies Running Local AI Models on Mac Without Cloud Dependency

Technology4 sources

NEAR AI Breach Exposes Critical Flaw in SGLang Security, Compromises 100 Users and 18 AI Agents

A remote code execution exploit in SGLang, an open-source inference engine, breached NEAR AI's systems on November 8. The attacker accessed account details for approximately 100 users and extracted chat history, settings and logs from 18 AI agents. The incident exposed limitations in hardware-backed privacy protections, as the Trusted Execution Environment failed to prevent the software vulnerability from compromising confidential AI requests.

Technology2 sources

NEAR AI Breach Exposes Critical Flaw in SGLang Security, Compromises 100 Users and 18 AI Agents

Technology2 sources

New York Establishes Nation's First Independent AI Safety Testing Center With 80 Researchers

New York Governor Kathy Hochul launched the nation's first state-run AI testing program, establishing the New York Center for AI Responsibility and Research at Binghamton University. Over 80 scientists will independently test AI systems for safety and security rather than relying on company self-regulation, with implementation tied to the RAISE Act's January 1, 2027 deadline.

Policy3 sources

New York Establishes Nation's First Independent AI Safety Testing Center With 80 Researchers

Policy3 sources

Prev12 3 4…99 Next