Tree Search for Web Agents: How tree search improves AI agents’ ability to browse the web and complete tasks
AI Agents

Tree Search for Web Agents: How tree search improves AI agents’ ability to browse the web and complete tasks

Browsing the web to achieve a specific goal can be challenging for agents based on large language models and even for vision-language models that can process onscreen images of a browser.

February 12, 20253 min read
Agents Go Deep: OpenAI’s Deep Research agent generates detailed reports by analyzing web sources
AI Agents

Agents Go Deep: OpenAI’s Deep Research agent generates detailed reports by analyzing web sources

OpenAI introduced a state-of-the-art agent that produces research reports by scouring the web and reasoning over what it finds.

February 12, 20252 min read
Agents Go Deep: OpenAI’s Deep Research agent generates detailed reports by analyzing web sources
AI Agents

Agents Go Deep: OpenAI’s Deep Research agent generates detailed reports by analyzing web sources

OpenAI introduced a state-of-the-art agent that produces research reports by scouring the web and reasoning over what it finds.

February 12, 20252 min read
Okay, But Please Don’t Stop Talking: Moshi, an open alternative to OpenAI’s Realtime models for speech
AI Agents

Okay, But Please Don’t Stop Talking: Moshi, an open alternative to OpenAI’s Realtime models for speech

Even cutting-edge, end-to-end, speech-to-speech systems like ChatGPT’s Advanced Voice Mode tend to get interrupted by interjections like “I see” and “uh-huh” that keep human conversations going. Researchers built an open alternative that’s designed to go with the flow of overlapping speech.

February 5, 20253 min read
Okay, But Please Don’t Stop Talking: Moshi, an open alternative to OpenAI’s Realtime API for Speech
AI Agents

Okay, But Please Don’t Stop Talking: Moshi, an open alternative to OpenAI’s Realtime API for Speech

Even cutting-edge, end-to-end, speech-to-speech systems like ChatGPT’s Advanced Voice Mode tend to get interrupted by interjections like “I see” and “uh-huh” that keep human conversations going. Researchers built an open alternative that’s designed to go with the flow of overlapping speech.

February 5, 20253 min read
Training for Computer Use: UI-TARS shows strong computer use capabilities in benchmarks
AI Agents

Training for Computer Use: UI-TARS shows strong computer use capabilities in benchmarks

As Anthropic, Google, OpenAI, and others roll out agents that are capable of computer use, new work shows how underlying models can be trained to do this.

February 5, 20253 min read
Training for Computer Use: UI-TARS shows strong computer use capabilities in benchmarks
AI Agents

Training for Computer Use: UI-TARS shows strong computer use capabilities in benchmarks

As Anthropic, Google, OpenAI, and others roll out agents that are capable of computer use, new work shows how underlying models can be trained to do this.

February 5, 20253 min read
Computer Use Gains Momentum: OpenAI’s Operator automates online tasks with a new AI agent
AI Agents

Computer Use Gains Momentum: OpenAI’s Operator automates online tasks with a new AI agent

OpenAI introduced an AI agent that performs simple web tasks on a user’s behalf.

January 29, 20252 min read
Computer Use Gains Momentum: OpenAI’s Operator automates online tasks with a new AI agent
AI Agents

Computer Use Gains Momentum: OpenAI’s Operator automates online tasks with a new AI agent

OpenAI introduced an AI agent that performs simple web tasks on a user’s behalf.

January 29, 20252 min read
Mustafa Suleyman: Agents of action
AI Agents

Mustafa Suleyman: Agents of action

In 2025, AI will have learned to see, it will be way smarter and more accurate, and it will start to do things on your behalf.

January 1, 20252 min read
Mustafa Suleyman: Agents of action
AI Agents

Mustafa Suleyman: Agents of action

In 2025, AI will have learned to see, it will be way smarter and more accurate, and it will start to do things on your behalf.

January 1, 20252 min read
Agents Ascendant: LLMs evolve with agentic workflows, enabling autonomous reasoning and collaboration
AI Agents

Agents Ascendant: LLMs evolve with agentic workflows, enabling autonomous reasoning and collaboration

The AI community laid the foundation for systems that can act by prompting large language models iteratively, leading to much higher performance across a range of applications.

December 25, 20242 min read
Agents Ascendant: LLMs evolve with agentic workflows, enabling autonomous reasoning and collaboration
AI Agents

Agents Ascendant: LLMs evolve with agentic workflows, enabling autonomous reasoning and collaboration

The AI community laid the foundation for systems that can act by prompting large language models iteratively, leading to much higher performance across a range of applications.

December 25, 20242 min read
Competitive Performance, Competitive Prices: Amazon introduces Nova models for text, image, and video
AI Agents

Competitive Performance, Competitive Prices: Amazon introduces Nova models for text, image, and video

Amazon introduced a range of models that confront competitors head-on.

December 11, 20244 min read
Competitive Performance, Competitive Prices: Amazon introduces Nova models for text, image, and video
AI Agents

Competitive Performance, Competitive Prices: Amazon introduces Nova models for text, image, and video

Amazon introduced a range of models that confront competitors head-on.

December 11, 20244 min read

Subscribe to The Batch

Stay updated with weekly AI News and Insights delivered to your inbox