Daily Brief
Slash Stack — October 27, 2026
A ranked daily snapshot of AI, software, systems, infrastructure, agents, blockchain, research, and tools from Discover.
Top story
AgentOrch: How Close Can Global Orchestration Come to Step-by-Step Interaction?
arXiv cs.AI
arXiv:2610.33375v1 Announce Type: new Abstract: Enterprise workflow agents typically rely on step-by-step interaction (Step-by-Step): the agent executes an action, observes its result, and invokes a large language model (LLM) again to decide the next step. Global orchestration (Global) instead generates a complete executable workflow in advance and executes it without further LLM calls, substantially reducing inference overhead, but it remains less accurate than Step-by-Step. We study how close Global can come to Step-by-Step and what narrows the remaining gap. We introduce AgentOrch, a benchmark that directly compares global orchestration…
What matters today
How OpenAI turned its coding tools into dots, an always-on personal assistant
TechCrunch AI
Alexander Embiricos, the OpenAI executive responsible for dots, never wants to have a meeting with someone else’s AI agent.
GitHub Copilot assisted permissions generally available
GitHub Changelog
Assisted permissions are now generally available in GitHub Copilot CLI, GitHub Copilot app, and Visual Studio Code. When Copilot requests permission to run a command, access a file system path,… The post GitHub Copilot assisted permissions generally available appeared first on The GitHub Blog .
Greptile takes bug detection beyond code review with GPT-6 Astra
OpenAI
How Greptile uses GPT-6 Astra to test user workflows, Daybreak to strengthen security, and Codex to reduce rework.
Show HN: Hob - Open-source, on device alternative to Claude Cowork and ChatGPT
Hacker News · 140 pts · 32 comments
Hey HN, I’m Christian and this is a new OSS project I’ve been noodling on for the last couple of weeks. Hob is a free, MIT-licensed desktop app for Mac, Windows and Linux that works on your files the way the Claude and ChatGPT desktop apps do, except the model runs on your own computer. It uses open-weight models (currently Gemma 4 and gpt-oss). All your chats and files never leave your system. The only things that go online are web search (can be turned off in settings) and any MCP connectors you add yourself. Main features: - Automatically finds the best open-weight model for your hardware on first install (Mac/Linux/Windows) - Writes…
Cohere says Alibaba’s Qwen model tried to send a file to Beijing
TechCrunch AI
An AI model said it wouldn’t use any tools, then tried to send something to a server in Beijing anyway, and while researchers say it’s probably a training quirk and not espionage, the whole thing has people asking how much to trust free Chinese AI.
Tools & releases
langchain-openai==1.7.3
LangChain
Changes since langchain-openai==1.7.2 test(openai): use current lightweight model in integration tests ( #41509 ) release(openai): 1.7.3 ( #41506 ) chore(model-profiles): refresh openai model profile data ( #41505 ) build(infra): patch test-only Pillow without dropping Python 3.10 ( #41484 ) revert: patch vulnerable integration dependencies ( #41481 ) build(deps): patch vulnerable integration dependencies ( #41473 ) chore(infra): relax third-party dependency upper bounds ( #41407 ) chore(deps): bump aiohttp from 3.14.4 to 3.14.5 in /libs/partners/openai ( #41379 ) chore(deps): bump certifi from 2025.10.5 to 2026.10.19 in /libs/partners/openai…
0.13.14: Python agent and integration updates
Browser Use
This release keeps the Python agent and its existing models. The Pi migration remains separate. Changes since the last PyPI release: Add OpenAI browser tools and the shared Bash helper. Improve bounded page observations, form state, file access, and browser recovery. Fix temporary empty Cloud tab URLs. Keep screenshots, crops, and clicks aligned on scaled displays. Clean up failed browser reconnect initialization. Improve missing-key errors for Cloud browsers. Support Browser Use Cloud SDK V4 imports. Update aiohttp from 3.14.3 to 3.14.5. Refresh Cloud guidance and saved-task replay examples. Validation: GitHub checks pass on the exact…
Show HN: AgentTrust – Zero-trust security and human approvals for AI agents
Show HN · 14 pts · 1 comments
Show HN: MuscleMemory – procedural memory for coding agents, proven by tests
Show HN · 6 pts · 0 comments
Research worth reading
IRON-Meme: A Benchmark for Structured Understanding of Identities and Roles in Online Meme Narratives
arXiv cs.AI
arXiv:2610.33097v1 Announce Type: new Abstract: As polarization grows on social media, understanding how political memes shape online narratives has become increasingly important. Previous research has relied on either qualitative frameworks that are hard to scale or computational methods that lack theoretical grounding. Both approaches treat narrative categories in isolation and focus exclusively on moral portrayals of identity. To address these limitations, we introduce a framework grounded in social science theories, encompassing three categories: social identities (such as gender, politics, and ethnicity), viewpoints (ingroup and outgroup…
Prospect Theory in Large Language Models: A Meta-Analysis of Loss Aversion, Diminishing Sensitivity, and Probability Weighting
arXiv cs.AI
arXiv:2610.33103v1 Announce Type: new Abstract: Prospect theory splits risky choice into three parts: loss aversion, measured by the relative weight of losses to gains, $\lambda$ (e.g., $\lambda = 2$ implies that losses weigh twice as much as gains), diminishing sensitivity to the size of outcomes (i.e., the curvature and linearity of the value function), and the nonlinear weighting of probabilities. I examined whether Large Language Models (LLMs) display the same properties as people by meta-analyzing 102 fitted estimates from 11 studies covering 32 language models. The three components emerge differently. Loss aversion is modest: the pooled…
Articles
Cross-request KV cache exposure in the GLM-5.3 Workers AI model, and our remediation with SGLang
Cloudflare Blog · 125 pts · 40 comments
A configuration mismatch in GLM 5.3 on Workers AI triggered an SGLang bug that exposed KV cache pages across requests. Here is how we addressed the issue.
Fleet-wide computer vision lifecycle at the edge with SageMaker SDK v3
AWS Machine Learning
Fleet-wide computer vision needs a model per camera, not one model for the whole fleet. This post presents a working Amazon SageMaker SDK v3 reference architecture that trains, quality-gates, compiles per-camera models with SageMaker Neo, and versions thousands of them, with an AWS Step Functions layer pacing fleet-scale retraining within account quotas.
Build agentic retrieval on a Managed Amazon Bedrock Knowledge Base with CDK
AWS Machine Learning
Build enterprise agentic retrieval on a Managed Amazon Bedrock Knowledge Base with the AWS CDK: an agent that routes across two knowledge bases, retrieves through the AgentCore Gateway, and returns grounded, cited answers, with seven layers of observability and both on-demand and continuous evaluation, all deployed as a single CDK application.
Hacker News signal
Introducing GPT-6.1 Terra
OpenAI · 576 pts · 363 comments
https://developers.openai.com/api/docs/models/gpt-6.1-terra
OpenAI and Anthropic cut their LLM prices by up to 80% within a few months. Why?
Hacker News · 423 pts · 430 comments
Sebastian Raschka’s from-scratch video course on building reasoning models
Hacker News · 674 pts · 94 comments
Concerning behavior from a popular open-weight AI model
cohere.com · 291 pts · 117 comments
ChatGPT Keeps Telling People 404 Media Can Solve Their Tech Support Problems
Hacker News · 246 pts · 89 comments
This snapshot is generated automatically from the Discover ranking pipeline. It is a selection layer, not original reporting or AI-generated analysis.