AI TodayGet the app

Agents

Curated AI news · 24 stories

AI Research

OpenAI’s first hardware device is reportedly a screenless speaker that can move

OpenAI is reportedly developing its first hardware device, a screenless mobile smart speaker with AI capabilities that can sync with ChatGPT and act as a personalized home companion. The device is said to have mechanical elements that allow it to move and learn about its owner over time, aiming to provide a unique interactive experience. This development comes amid legal disputes with Apple, which has accused OpenAI of trade secret theft, though OpenAI denies these claims.

Agents

Multi-agent social intelligence with Strands Agents and Amazon Bedrock

The article discusses how Thrad.ai utilizes multi-agent systems built with Strands Agents and Amazon Bedrock AgentCore to automate social intelligence gathering from diverse online sources for targeted advertising. It explains the orchestration of specialized agents to analyze signals across platforms and generate personalized outreach emails, comparing different orchestration patterns and detailing scoring and governance methods. The approach is applicable to various fields like competitive and

Agents

Hermes agent maker Nous Research in talks for new funding at $1.5B valuation

Nous Research, the developer of the open-source Hermes AI agent, is closing a funding round led by Robot Ventures at a $1.5 billion valuation, raising at least $75 million. Hermes is a versatile AI agent with built-in skills for tasks like web search and coding, available both as open-source software and a cloud-hosted service. The new funds aim to support further product development and business expansion.

Healthcare

Building an agentic AI solution at Bluesight with Amazon Bedrock

Bluesight, a healthcare software company, developed an AI solution called Prism using Amazon Bedrock to streamline compliance and inventory management across multiple hospital systems. The AI agent integrates data from various sources to reduce manual audit work and provide actionable insights, with initial deployment in 20 health systems and plans for broader use. This approach addresses complex compliance challenges by automating data analysis and reporting across several healthcare products.

Agents

Implement on-behalf-of token exchange for multi-tenant agents with Amazon Bedrock AgentCore Gateway

The article explains how Amazon Bedrock AgentCore Gateway implements OAuth 2.0 Token Exchange (RFC 8693) to enable multi-tenant generative AI agents to securely propagate user identity when calling downstream APIs. It details a reference implementation called TravelBot that demonstrates on-behalf-of token exchange to maintain fine-grained access control and scalable identity management across multiple tenants. This approach addresses challenges in multi-tenant architectures where agents act on a

Agents

Cursor is building a Claude Cowork rival, and Musk may decide its fate

Cursor, known for its AI tools for software developers, is developing Sand, a general-purpose AI agent aimed at office workers to compete with Anthropic's Claude Cowork and OpenAI's ChatGPT Work. Sand integrates deeply with various platforms and extends Cursor's reach beyond coding to broader office tasks, but its launch depends on computing resources leased from Elon Musk's SpaceXAI. This development adds a new competitor in the race to automate office workflows with AI agents.

Agents

Scaling agentic workflows with native case management in Amazon Quick Automate

Amazon Quick Automate integrates native case management to enable scalable, agent-driven automation of complex business workflows. This approach tracks each work item as a case through its lifecycle, supports human intervention when needed, and allows dynamic scaling and exception handling for enterprise processes. The article details how case management enhances visibility, control, and reliability in automating end-to-end business operations.

Agents

How KTern.AI built agentic AI for SAP on Amazon Bedrock AgentCore

KTern.AI leveraged Amazon Bedrock AgentCore to develop autonomous AI agents that manage complex SAP digital transformation workflows, enhancing automation without the need for custom infrastructure. These agents handle tasks such as reverse engineering and exception mining across finance and sales, enabling faster and more reliable enterprise-scale SAP migrations and conversions. The platform integrates domain expertise and data-driven automation to improve transformation speed and reduce effort

AI Research

Malaysia’s PM is sending an autonomous AI double out to serve citizens, payment links and all

Malaysia's Prime Minister Anwar Ibrahim is set to launch an autonomous AI avatar named PMX AI, designed to interact with citizens by performing tasks such as renewing driving licenses and providing career advice. Developed by Zetrix AI Bhd., the avatar mimics Anwar's voice and mannerisms, supports multiple languages and dialects, and aims to enhance government services ahead of the upcoming national election. This initiative represents a novel use of AI in governance, extending beyond simple AI-

AI Research

OpenAI is killing its ChatGPT Atlas browser less than a year after launch

OpenAI is discontinuing its ChatGPT Atlas AI-powered web browser less than a year after its October 2025 launch, with plans to retire it by August 9. Instead of abandoning browsing capabilities, OpenAI is integrating Atlas's features into a new unified desktop app called ChatGPT Work, which combines various tools including a built-in browser and cloud-based agents to perform tasks. This consolidation reflects OpenAI's strategy to streamline its product offerings and focus on productivity amid a競

Agents

An AI agent startup just let its agent run its $100M fundraise

Lyzr, a startup specializing in AI agents, utilized its own AI system, SivaClaw, to manage and execute its $100 million Series B fundraising round, engaging over 130 investors and handling tasks like drafting memos and tracking investor interest. This approach demonstrated the product's capabilities and highlighted a trend where AI startups can attract significant investment without traditional in-person efforts. The company garnered $400 million in investor interest globally without founders'需需

Agents

OpenAI is shutting down Atlas, but its AI browser ambitions are still growing

OpenAI is discontinuing its AI-powered browser Atlas but continues to integrate its browsing features into the ChatGPT desktop app and a new Chrome extension. This strategy reflects OpenAI's view of the browser as a feature rather than a standalone destination, aiming to enhance user experience within existing platforms. The updates enable ChatGPT to function as a continuous workspace across multiple environments, including Chrome and its desktop app.

LLM & Text Generation

OpenAI rolls out GPT-5.6 after government greenlight — and announces ‘ChatGPT Work’

OpenAI has received government approval to publicly release its GPT-5.6 model, which was previously available only to select organizations. Concurrently, OpenAI introduced ChatGPT Work, an AI agent combining ChatGPT and Codex functionalities to assist non-technical users with tasks like document creation and app integration. The rollout is ongoing globally, with varying access depending on platform and subscription level, as OpenAI aims to compete with similar AI agents in the market.

Agents

Lyzr used its own AI agent to help raise a $100mn round

Lyzr, an enterprise AI agents startup, utilized its own AI agent named Agent Sam to assist in raising a $100 million Series B funding round, which could value the company at around $500 million. Agent Sam managed investor communications and drafted investment memos, demonstrating the company's product capabilities while attracting significant investor interest. Founded in 2023, Lyzr focuses on providing AI agent software with strong governance features for regulated industries, positioning its  

Agents

ChatGPT is now a partner for your most ambitious work

OpenAI has introduced ChatGPT Work, an advanced agent powered by the GPT-5.6 model and Codex technology, designed to assist users with complex, multi-step tasks across various applications. This tool can autonomously manage workflows, create documents, and update materials while integrating information from multiple sources, aiming to enhance productivity in areas like sales and finance. Early internal use at OpenAI demonstrates its capability to accelerate processes and handle tasks that would,

Agents

Jensen Huang says his engineers would rather build agents than write code

Nvidia CEO Jensen Huang stated that his software engineers now prefer creating AI agents over traditional coding, viewing this shift as an advancement in their work. These AI agents handle routine coding tasks, allowing engineers to focus on designing, benchmarking, and ensuring system safety. Huang envisions widespread use of AI agents within Nvidia to boost productivity and believes AI will create jobs rather than eliminate them.

LLM & Text Generation

The Future Is Always Listening: OpenAI Says Its New Voice Assistant Is ‘One Step Closer to a Truly Accessible AGI’

OpenAI has introduced GPT-Live-1, a new AI voice assistant designed to engage in more natural and humanlike conversations by incorporating speech nuances and knowing when to listen quietly. The model can handle complex requests by deferring to GPT-5.5 and includes enhanced real-time web search and translation features, aiming to serve as a versatile and accessible AI companion. This development represents a step toward more advanced and accessible artificial general intelligence (AGI).

LLM & Text Generation

OpenAI releases new voice models for more natural live conversations

OpenAI has introduced new conversational voice models, GPT-Live-1 and GPT-Live-1 mini, which enable more natural, full-duplex interactions allowing users to speak and listen simultaneously. These models improve turn-taking, support longer conversations, and integrate with advanced text models like GPT-5.5 for enhanced reasoning and visual responses. The updated voice mode replaces the previous system in ChatGPT and aims to make voice a primary interface for complex computing tasks.

Agents

Prime Intellect raises $130M Series A to help enterprises build their own AI agents

Prime Intellect, a startup founded in 2024, has raised $130 million in a Series A funding round to provide enterprises with computing power and software tools for building AI agents. Their platform offers a modular, full-stack solution including reinforcement learning frameworks, enabling companies to develop custom AI systems without relying on external AI labs. The company has quickly gained customers and reached a $100 million annualized revenue run rate by delivering efficient, cost-saving,,

Agents

Amazon’s secret ‘Moonraker’ project pushes Alexa into the AI-agent race

Amazon is developing a new project called Moonraker to enhance Alexa's capabilities, enabling it to perform complex, multi-step tasks from a single command. This initiative aims to compete with AI agents from other tech companies but comes with high costs, including projected GPU expenses exceeding $100 million in 2026. Despite challenges such as delays and performance issues, Amazon continues to invest in improving Alexa's functionality.