Travelers has implemented a new AI tool developed with OpenAI to assist customers with the claims process. This system provides round-the-clock support and helps the company manage high volumes of requests.
Read the source →Anthropic is extending the scope of Project Glasswing. This initiative is currently being expanded.
Read the source →Holo3.1 is a new model designed for local computer use agents. It focuses on providing high-speed performance for automating on-device tasks.
Read the source →NVIDIA is partnering with industrial software firms to develop autonomous AI engineers using the NemoClaw framework. The goal is to automate complex simulation workflows, including design, setup, and report generation.
Read the source →AWS provides a guide on balancing domain-specific fine-tuning with the preservation of a model's general abilities. The resource details how to manage key training variables and avoid costly errors during the customization process.
Read the source →Uber has placed a cap on staff spending for artificial intelligence. This move follows a period where employees were urged to maximize AI usage, leading the company to exhaust its budget in just four months.
Read the source →CEO Satya Nadella unveiled several updates, including new Surface devices and a persistent personal AI assistant. The event also featured enhancements to the company's internal AI models.
Read the source →Microsoft has launched Scout, an AI coworker integrated directly into Teams. The agent is designed to automate repetitive office chores and operate like a human teammate.
Read the source →The worldwide healthcare system is facing severe pressure due to long-term underfunding and labor shortages. This crisis is worsened by an aging population and rising staff burnout.
Read the source →OpenAI is adding new plugins tailored for investment banking, sales, and data analysis to Codex. With non-developers joining the platform three times faster than coders, the company aims to turn the tool into a general productivity application.
Read the source →Microsoft is working on Project Solara, a new version of Android. This operating system focuses on AI agents rather than traditional applications.
Read the source →OpenAI has introduced new annotations, sites, and plugins for Codex. These updates are designed to increase productivity for various professional roles, including marketers and analysts.
Read the source →The two companies are launching a unified software and hardware stack to support the deployment of AI agents. This integration will be available across Azure cloud services, local environments, and Windows hardware.
Read the source →AWS has provided a guide on using Amazon Nova 2 Lite for identifying objects within images. The tutorial demonstrates how to integrate the model with Bedrock, Lambda, and API Gateway for use in sectors like logistics and farming.
Read the source →Microsoft has introduced an open-source tool called Adaptive Spec-driven Scoring for Evaluation and Regression Testing. This framework allows developers to create AI behavior tests using text-based descriptions.
Read the source →President Trump has established a voluntary system for AI developers to share new frontier models with the government prior to public launch. The initiative aims to enhance critical infrastructure cybersecurity and support safe innovation.
Read the source →OpenAI has detailed its priorities for AI governance and public policy. The framework focuses on safety, global norms, protecting young people, and managing job market shifts.
Read the source →Google is integrating AI capabilities into its Search and Shopping platforms to improve the vintage shopping experience. These tools are designed to help users more easily find unique pre-owned items.
Read the source →New research explores applying Direct Preference Optimization to use cases outside of conversational AI. This work examines how the technique can be adapted for various other AI applications.
Read the source →NVIDIA Research is developing new methods to improve how robots handle unfamiliar objects. The work also focuses on enhancing the safety of self-driving vehicles and the efficiency of training AI agents.
Read the source →Fundamental's large-scale model for tabular data is now accessible via Amazon SageMaker JumpStart. Users can deploy the model and use it to generate predictions from corporate datasets.
Read the source →Amazon is implementing AI and visual search tools to display generated imagery based on user queries. The company believes this feature will better assist customers in finding specific items.
Read the source →Reviewers found Google's new Spark agent to be alarmingly capable. The tool demonstrated the ability to access personal details about the users' families and pets without being told.
Read the source →Donald Trump has officially signed an executive order regarding artificial intelligence. This action follows a period where the initial proposal had been set aside.
Read the source →A new White House directive tasks agencies such as CISA and the Pentagon with enhancing cyber defenses using AI. While the order allows developers to submit their models for security evaluations, it specifies that these reviews are not mandatory.
Read the source →Meta is working to narrow the distance between its technology and that of its competitors. However, uncertainty remains regarding the company's ability to achieve parity.
Read the source →OpenAI has introduced a suggested federal framework for managing advanced AI systems in the United States. The plan focuses on enhancing national security, safety, and overall resilience.
Read the source →The Reachy Mini robot has been updated to support MCP tools. This integration expands the robot's functional capabilities.
Read the source →NVIDIA has presented new agent skills at CVPR to accelerate the creation of robotics, vision AI, and self-driving technology. These tools focus on improving the entire development workflow, including scene reconstruction and the generation of edge-case scenarios.
Read the source →AWS is implementing SOCI indexes to decrease cold start latency for Deep Learning AMIs and containers. The company provides guidance on selecting the appropriate SOCI modes to enhance workload efficiency.
Read the source →Former Meta and Goldman Sachs employees have launched a voice AI company focusing on the Middle East and Africa. Their current technology manages over 17,000 daily calls.
Read the source →Amazon's search tool now creates AI-generated images of home decor and apparel based on user descriptions. Customers can use these synthetic visuals to find similar available merchandise.
Read the source →