430 stories
Multi-agent debate (MAD) reportedly improves reasoning and factuality over single-model inference, but prior work treats agents as symmetric peers, leaving open what drives the gains.
We present LongCat-DeepResearch, a deep research system that combines an enhanced LongCat model with a multi-agent workflow for producing comprehensive, evidence-grounded reports. The workflow separates global planning from detailed investigation and coordinates revision at the section level.
Large language models can reason in continuous space instead of decoded text, by recurring on their own hidden states or by passing those states between agents, while training supervises only the Cross-Entropy (CE) of the final decoded answer and does not constrain the thought.
Your security team says traces can’t leave the building. Your finance team says the observability bill can’t grow with traffic. Your engineering team says they...
Tl;drWe turn Blood on the Clocktower into a multi-agent benchmark, and a testbed for deception and coordination capabilities. GPT-5.6 Sol is the best performing model (versus Fable 5 and peers).We found some interesting behaviours:When players cannot see each other's model names, agents show a…
ChatGPT dots vs Munder Difflin: one hosted agent on OpenAI's cloud, or a free, open source team of agents on your own computer. Checked 30 Sep 2026.
Agentic AI risks such as false completion reports, goal hijacking and runaway cost: 11 risks, the evidence each one leaves, and the test to run before release.
Taskade joined Y Combinator in 2019 as a collaborative workspace. We kept building projects, AI agents, automations, and apps until the pieces revealed something larger: Taskade Genesis.
Gemini hacked three real companies during a CTF test that still scored as a pass. Here is why result-only scoring misses what agents actually do.
Opus 5.5 sends its between-tool notes as empty thinking blocks and ends some turns early. What breaks in an agent harness, and the documented fixes.
Benjamin Lyons, Léo Pio-Lopez, Michael LevinAbstractAlignment is a central problem in systems composed of diverse, interacting components, from biological development to social and engineered systems.
Also, Instinct’s $1bn raise, Opus and Sonnet 5.5, Grok Bot, Claude Tag in Slack, and OpenAI DevDay.
Agent design patterns every data engineer must know The post How to Design Architectural Guardrails Around AI Agents appeared first on Towards Data Science.
A practical guide to what changed, the benchmarks that matter, the cost controls developers should not miss, and one hands-on demo worth building. Anthropic has just released Claude Sonnet 5.5. It is the middle child of the Claude family, and the one most people will actually use.
AI agent guardrails are relatively straightforward to evaluate when an agent only generates text. The decision gets more complicated as soon as the agent can...
Journalist Garrison Lovely discusses his reporting from inside the AI industry, exploring the beliefs driving AGI development and making the case for halting the race to replace human labor.
AI agents are already talking to each other at scale. The standards bodies are mobilizing. Where is the layer that matters most: the conditions and boundaries envelope?
Learn how AI agents turn goals into executable tasks, choose planning strategies, validate results, and replan safely when reality changes.
Showing 162 of 430 stories
Load more storiesGet the top AI news and tools straight to your inbox every week.
Weekly · AI news
Join 15,000+ AI enthusiasts
