Translate
Live intelligence

Latest AI updates

93 updates
AllModelsProductsIndustryResearchTutorialOpinion
Research Editor selected

CWE-bench v2: More coverage, quality, and transparency

Hugging Face released CWE-bench v2, an updated benchmark for evaluating AI agents on real-world software vulnerabilities. The new version expands coverage to 25 CWE categories, improves data quality through manual curation, and adds transparency features like per-sample metadata and leaderboard filtering. This update aims to provide a more reliable and comprehensive evaluation for AI-driven security tools.

Why it matters: Provides concrete details on a new benchmark version, directly useful for AI security researchers and developers evaluating agent performance.

Hugging Face Blog2d agoScore 85
AI benchmarkssoftware securityAI agents
Industry Editor selected

Building AI for Lawyers: Legora CTO on Agents, Evals, and the End of the Billable Hour

Legora CTO Jacob Lauritzen discusses how the legal AI startup grew from a ChatGPT-with-Swedish-law prototype to a $8.8B company with $200M+ ARR. He explains how Legora evaluates agents without accessing confidential client data, using simulated environments and a team of legal engineers. The interview covers model selection, the shift from assistant to agent, and how the billable hour is being disrupted by flat fees and AI-driven efficiency.

Why it matters: Provides rare, detailed insight into how a leading legal AI company builds and evaluates agents at scale, with concrete metrics and product philosophy.

latent2d agoScore 92
Legoralegal AIAI agents
Industry

Clark raises $11M to put AI 'Clerks' to work on company audits

New York startup Clark raised an $11M seed round led by Touring Capital and launched Clerks, AI agents designed to work through audits for CPA firms. The funding and product launch signal growing interest in AI-driven automation for professional services.

AIHUB3d agoScore 78
AI in accountingventure capitalprofessional services automation
Industry

The agentic web is here

This Technology Review feature argues that LLM-based agents are beginning to reshape the web, moving from browser-automation experiments to tool connectors and assistant agents from Meta, OpenAI, and Anthropic. It describes early commercial responses, including agent-readable publishing and agent-targeted ads, alongside platform resistance such as Amazon blocking Meta's Muse. The piece also examines how agent traffic could disrupt advertising and publishing economics, and how control may concentrate around a few closed agent platforms.

technologyreview3d agoScore 68
AI agentsagentic webplatform economics
Products Editor selected

NVIDIA Releases Auto Ontology: An Open Semantic Layer That Grounds AI Agent SQL

NVIDIA has released Auto Ontology, an open-source library under the Apache 2.0 license. The library generates a semantic layer from enterprise metadata, designed to ground AI agent SQL queries in structured business context. This release aims to improve the reliability and accuracy of AI agents interacting with databases.

Why it matters: Provides a concrete, open-source tool from a major AI company that directly addresses a key challenge in enterprise AI agent deployment.

AIHUB3d agoScore 85
NVIDIAAuto Ontologysemantic layer
Research Editor selected

Operon-S1: a fast decision controller for long-horizon biomedical agents

Operon-S1 is a fast decision system that separates judging from doing in language-model agents for biology. It poses recurring agent judgments as closed questions at 13 decision points, using a local model that runs on CPU in 0.6 seconds with no request cost. The system records every decision with its consequence and can be retrained on labels from task construction, verifiers, or execution. It makes agent judgments a measurable, replaceable, and trainable layer without affecting task scores.

Why it matters: Introduces a practical architecture for transparent, fast, and trainable decision-making in biomedical AI agents, with measurable efficiency gains.

arxiv3d agoScore 88
AI agentsbiomedical AIdecision systems
Research Editor selected

BlindToChange: AI Agents Fail to Proactively Search for Recent Events

A new benchmark, BlindToChange, reveals that AI agents often fail to proactively search for recent real-world events that affect their responses, even though they can find changes when explicitly asked. The benchmark uses requests where an obvious candidate is invalidated by a recent event, and shows a performance gap of up to 27.5 points when the event is not mentioned. Over half of failures stem from agents not directly verifying relevant facts. This work highlights a new axis for progress in agentic systems.

Why it matters: Provides a concrete, measurable benchmark exposing a critical blind spot in AI agent behavior, directly relevant to improving real-world reliability.

arxiv3d agoScore 85
AI agentsbenchmarkproactive information seeking
Industry

SK Ecoplant Rolls Out Tailored AI Agents for Every Employee

SK Ecoplant is deploying tailored AI agents for every employee, beginning at semiconductor and AI data center sites. The company has trained 318 in-house AI experts to support the rollout.

AIHUB3d agoScore 72
SK EcoplantAI agentsenterprise deployment
Products

Google’s Computer Use Goes GA; Client Apps Still Execute the Actions

Google’s Computer Use feature has reached general availability across two Gemini APIs, according to the report. The item notes that client applications still execute model-suggested actions and continue to implement the screenshot, environment, and safety loop.

AIHUB4d agoScore 72
Google GeminiComputer UseAI agents
Research

Adobe Research and UC San Diego Build EditAgent for Professional-Style AI Image Editing

Adobe Research and UC San Diego developed EditAgent, an AI system that edits images by selecting specialized tools for each task rather than repeatedly invoking a general-purpose editor. A standard language model chooses the best result based on measured image data. The approach works but requires about ten times the compute.

The Decoder4d agoScore 74
Adobe ResearchUC San DiegoAI agents
Products

DeepSeek unveils AI 'super agent' in coding race with OpenAI, Anthropic

DeepSeek has previewed a plan for a more autonomous AI agent designed to handle multiple tasks with minimal human intervention, intensifying competition with OpenAI and Anthropic in the coding assistant space. The announcement signals a strategic push toward agentic AI capabilities.

AIHUB4d agoScore 75
DeepSeekAI agentscoding assistants
Products

DeepSeek Open-Sources dsh-superagent, a Router-Worker Coding Agent

DeepSeek's Harness team released dsh-superagent, an MIT-licensed research prototype that acts as a proactive coding agent for GitHub repositories. It uses a router-worker architecture where a long-running router dispatches tasks to isolated worker sessions, each handling a single ticket. In simulations, it achieved a p95 task completion time of 15.7 minutes versus 29.5 minutes for a single-session baseline, though real coding quality was not measured. The agent is designed to minimize human attention, with proactive work gated by an observe mode by default.

MarkTechPost4d agoScore 78
DeepSeekAI agentscoding automation
Industry Editor selected

Arca Builds Memory and Safety Systems for Long-Horizon Wealth Management Agents

Arca uses AI agents to support human financial advisors on tasks that can span up to ten days, with an internal memory system tracking client work, handoffs, responsibilities and next steps. The company built a custom model-selection harness using internal benchmarks and applies deterministic restrictions so agents cannot execute trades or move money. Arca also uses constrained data-querying formats, prompt-injection checks and advisor feedback to improve its agent infrastructure.

Why it matters: The article provides concrete implementation details about deploying AI agents safely and reliably in a regulated financial-services workflow.

latent5d agoScore 86
ArcaAI agentsfinancial services
Products Editor selected

Vercel Labs Releases Open-Source Desktop Control Tool for AI Agents

Vercel Labs has released agent-desktop, an Apache-2.0-licensed CLI and C library that lets AI agents operate apps on macOS, Windows and Linux through operating-system accessibility trees rather than screenshots. The tool returns structured JSON and requires no embedded LLM or vision model, with accessibility-based actions designed to avoid moving the pointer or taking focus. Its reported token savings are self-reported, lack independent benchmarking and do not address apps without accessible UI trees.

Why it matters: The release gives developers a concrete, model-agnostic option for desktop automation while exposing clear accessibility constraints and unverified efficiency claims.

MarkTechPost5d agoScore 85
Vercel LabsAI agentsComputer use
Industry

DoorDash: External AI Agents Have Driven Only Hundreds of Orders

On DoorDash's third-quarter 2026 earnings call, co-founder and CEO Tony Xu said outside AI agents have generated "several hundred orders" in total since launching. He called it a "very low use case" for a platform that handled just under 1 billion orders in the quarter alone. Xu said agents may prove genuinely useful for long-consideration purchases and for proactive background work on a user's behalf, but simple tasks like ordering food still go directly to DoorDash.

Ars Technica AI5d agoScore 74
DoorDashAI agentsagentic commerce
Research Editor selected

New Evaluation Separates AI Agents’ Conversation Retrieval, Re-engagement and Continuity

Researchers introduce a framework for evaluating whether LLM agents can find inactive conversations, autonomously return to them and reconstruct enough context for coherent continuation. Across six LLM configurations, conversational resurfacing improves target conversation identification, while persistent memory is associated with more autonomous target-directed replies. Neither produces consistent improvements in contextual reconstruction, suggesting that retrieval and re-engagement need to be evaluated separately from conversational continuity.

Why it matters: The study gives agent developers a concrete framework for testing long-term conversational capabilities that memory retrieval metrics alone may miss.

arxiv6d agoScore 83
AI agentsLong-term memoryAgent evaluation
Research Editor selected

Hybrid Agent Memory Combines Episodic Recall with Structured Relationships

A new arXiv paper proposes a long-term memory architecture for dialogue agents that combines episodic memories, structured relational memories and a consolidation mechanism designed to preserve efficiency. The authors report that the memory types complement each other, achieving 90.2% accuracy on LoCoMo and 85.5% on LongMemEval-S. They report outperforming state-of-the-art memory frameworks on these evaluations.

Why it matters: The paper offers a concrete architecture and benchmark evidence for improving agents’ recall of facts, events and relationships across long interaction histories.

arxiv6d agoScore 83
AI agentsLong-term memoryMemory consolidation