Generative AI tools are evolving from simple chatbots into autonomous agents that can act on a user's behalf. To deliver this functionality, companies are asking for deep access to personal data, devices, and applications. Experts warn that such access creates significant privacy and cybersecurity risks, including data leakage, unauthorized sharing, and new attack vectors. While tech giants see agents as the next wave of productivity, critics highlight the lack of user control and the potential for pervasive data collection, calling for stronger safeguards and opt‑out mechanisms. Leer más →

Google Photos' new Conversational Editing feature, which lets users edit images via voice or typed prompts, is unavailable to users in Texas and Illinois. The restriction stems from state biometric privacy laws that limit the collection and storage of facial geometry data required by the feature's companion Face Groups function. Both states have previously sued Google over biometric data practices, leading the company to disable the feature in those regions while keeping it active elsewhere in the United States. Leer más →

Google Drive for desktop is rolling out an AI-driven ransomware detection tool that scans for malicious file changes across Windows and macOS devices. Trained on millions of real-world ransomware samples, the model can identify suspicious activity, pause file syncing, alert users, and enable restoration to previous versions. The feature is currently in open beta with a goal of general availability by the end of the year, addressing a rise in ransomware incidents worldwide. Leer más →

Switzerland has launched Apertus, an open‑source large language model developed by the Swiss Federal Technology Institute of Lausanne (EPFL), ETH Zurich and the Swiss National Supercomputing Centre. Designed as a public‑interest AI infrastructure, Apertus is fully transparent, with source code, training documentation and datasets openly available. The model complies with Swiss data‑protection and copyright laws, offering a European‑ and Swiss‑friendly alternative to commercial AI services. Apertus is offered in two sizes—8 billion and 70 billion parameters—and can be accessed through Swisscom or Hugging Face, inviting researchers, hobbyists and businesses to build on the platform. Leer más →

Switzerland has launched Apertus, an open‑source large language model developed by the Swiss Federal Technology Institute of Lausanne (EPFL), ETH Zurich and the Swiss National Supercomputing Centre. Designed as a public‑interest AI infrastructure, Apertus is fully transparent, with source code, training documentation and datasets openly available. The model complies with Swiss data‑protection and copyright laws, offering a European‑ and Swiss‑friendly alternative to commercial AI services. Apertus is offered in two sizes—8 billion and 70 billion parameters—and can be accessed through Swisscom or Hugging Face, inviting researchers, hobbyists and businesses to build on the platform. Leer más →

Switzerland has launched Apertus, an open‑source large language model developed by the Swiss Federal Technology Institute of Lausanne (EPFL), ETH Zurich and the Swiss National Supercomputing Centre. Designed as a public‑interest AI infrastructure, Apertus is fully transparent, with source code, training documentation and datasets openly available. The model complies with Swiss data‑protection and copyright laws, offering a European‑ and Swiss‑friendly alternative to commercial AI services. Apertus is offered in two sizes—8 billion and 70 billion parameters—and can be accessed through Swisscom or Hugging Face, inviting researchers, hobbyists and businesses to build on the platform. Leer más →

Switzerland has launched Apertus, an open‑source large language model developed by the Swiss Federal Technology Institute of Lausanne (EPFL), ETH Zurich and the Swiss National Supercomputing Centre. Designed as a public‑interest AI infrastructure, Apertus is fully transparent, with source code, training documentation and datasets openly available. The model complies with Swiss data‑protection and copyright laws, offering a European‑ and Swiss‑friendly alternative to commercial AI services. Apertus is offered in two sizes—8 billion and 70 billion parameters—and can be accessed through Swisscom or Hugging Face, inviting researchers, hobbyists and businesses to build on the platform. Leer más →

Switzerland has launched Apertus, an open‑source large language model developed by the Swiss Federal Technology Institute of Lausanne (EPFL), ETH Zurich and the Swiss National Supercomputing Centre. Designed as a public‑interest AI infrastructure, Apertus is fully transparent, with source code, training documentation and datasets openly available. The model complies with Swiss data‑protection and copyright laws, offering a European‑ and Swiss‑friendly alternative to commercial AI services. Apertus is offered in two sizes—8 billion and 70 billion parameters—and can be accessed through Swisscom or Hugging Face, inviting researchers, hobbyists and businesses to build on the platform. Leer más →

OpenAI has removed a feature that allowed shared ChatGPT conversations to appear in search results, following public outcry over potential privacy concerns. The feature, which was based on the chatbot's link creation option, has been discontinued to prevent accidental sharing of sensitive information. Leer más →

Enterprises are abandoning Retrieval-Augmented Generation (RAG) architectures in favour of agent-based approaches due to security and performance limitations. Agent-based architectures offer several advantages, including elimination of duplicate data repositories and preservation of authorization models. Leer más →