Brief diario
Slash Stack — 9 de octubre de 2026
Snapshot diario priorizado de IA, software, sistemas, infraestructura, agentes, blockchain, research y herramientas desde Discover.
Historia principal
Sophos cuts threat investigation time by 96% with OpenAI Daybreak
OpenAI
Discover how Sophos uses OpenAI’s Daybreak to cut cyber-threat investigation time by 96% and automate 52% of MDR cases while preserving human oversight.
Lo que importa hoy
Show HN: Jevman – AI decision models play Pac-Man
Hacker News · 61 pts · 14 comentarios
Openai just launched their decisions endpoint, cloudflare launched clef the other week, and many more jev alternatives are out there. We wanted to put the popular ones to the test and thought Pac-Man is a good benchmark for simple and fast decision making. So we let jev 1.13, kev, clef, clef flash, GPT-6 Luna and Laya play Pac-Man against bot ghosts. The low latency of these models allows for real time play. We had each model play 100 games, published a leader board and open-sourced the repo so anyone can run their own model and join the ranking. Link to repo: https://github.com/opper-ai/jevman-benchmark/blob/main/CONTR… You can also join…
Anthropic changes usage policy to ban model abuse and election interference
TechCrunch AI
Anthropic’s updated usage policy explicitly prohibits users from repeatedly abusing Claude in extreme cases, though ordinary frustration and criticism are still allowed. The new rules also address election interference, deceptive campaigns, weapons software, and surveillance.
Google brings agentic AI to Gemini, starting with businesses
TechCrunch AI
Google is turning Gemini into an AI agent that can plan, execute tasks, and work across business apps and systems. The agent can delegate work to subagents, use multiple AI models, and even gets its own workplace identity, complete with an email address.
LegalOn halves Codex costs while maintaining development speed
OpenAI
LegalOn cut estimated daily Codex costs by 65% while maintaining development speed. It matched Astra, Sol, and Luna to tasks and managed budgets strategically.
How Oracle turns days of work into minutes with ChatGPT and Codex
OpenAI
Across recruiting, engineering, and operations, Oracle turns specialist knowledge into fast, repeatable workflows with ChatGPT Work and Codex.
Tools y releases
langchain==1.4.4
LangChain
Changes since langchain==1.4.3 release(langchain): 1.4.4 ( #41163 ) fix(langchain): (SummarizationMiddleware) retry summary step on context overflow ( #41159 ) chore(deps): bump fsspec from 2026.6.0 to 2026.9.0 in /libs/langchain_v1 ( #41097 ) chore(deps): bump multidict from 6.7.0 to 6.9.1 in /libs/langchain_v1 ( #41075 ) chore(deps): bump langgraph-sdk from 0.4.2 to 0.4.4 in /libs/langchain_v1 ( #41074 ) chore(deps): bump fsspec from 2025.10.0 to 2026.6.0 in /libs/langchain_v1 ( #41071 ) chore(langchain): bump minimum FastMCP to 4.0.11 ( #41054 ) chore(deps): bump pyjwt from 2.13.0 to 2.15.0 in /libs/langchain_v1 ( #40939 ) chore(deps):…
langchain-openai==1.7.0
LangChain
Changes since langchain-openai==1.6.7 release(openai): 1.7.0 ( #41154 ) feat(openai): support decisions api ( #41125 ) chore(deps): bump langgraph-sdk from 0.4.4 to 0.4.6 in /libs/partners/openai ( #41096 ) chore(deps): bump langgraph-sdk from 0.4.2 to 0.4.4 in /libs/partners/openai ( #41072 ) chore(deps): bump pyjwt from 2.15.0 to 2.15.1 in /libs/partners/openai ( #40951 ) chore(deps): bump urllib3 from 2.7.0 to 2.8.0 in /libs/partners/openai ( #40952 )
Show HN: I Put an AI Agent on a Nokia 110
Show HN · 29 pts · 13 comentarios
I was using this mobile to reduce my screentime. Recently got the idea to put ai agent in it. Started with installing android, but failed as it has only 48MB RAM. Then reverse-engineered for few days and finally able use AI Agent in this and automated few actions. Thanks
Show HN: Aura – a self-hosted, multi-user AI agent with per-person graph memory
Show HN · 6 pts · 3 comentarios
Research que vale la pena leer
Whose Ground Truth? Embracing Ambiguity in Human-Centered AI
arXiv cs.AI
arXiv:2610.10805v1 Announce Type: new Abstract: As AI systems increasingly interact with people and make decisions about them, understanding human interpretations becomes an important part of developing human-centered AI. Conventional machine learning and AI systems are largely developed under the assumption that a single definitive ground truth exists, with variability in human annotations often resolved through aggregation or treated as noise. However, for many human-centered tasks, human interpretation is inherently ambiguous, and multiple interpretations of the same input may be simultaneously reasonable and valid. Reducing such ambiguity…
How Narrative Wrapping Affects LLM Refusal: A Cross-Language Benchmark and Defense
arXiv cs.AI
arXiv:2610.11005v1 Announce Type: new Abstract: Safety-aligned language models often refuse a harmful request stated directly but answer the same request inside a role-play or narrative wrapper. We measure this vulnerability across languages and registers: attack success on Qwen3-1.7B is already 89.4% in English and 93.0% in modern Chinese, and reaches 95.7% in Classical Chinese. We build GUISE, a benchmark for systematically studying this vulnerability. It includes parallel requests in English, modern Chinese, and Classical Chinese, matched harmful and benign pairs, wrapper types held out for evaluation, and a stricter criterion that counts…
Artículos
Roundtables: A Conversation With the Creator of AI-Designed Viruses
MIT Technology Review AI
Friday, October 16, 2026 Can AI design new life forms? In 2025, Stanford University PhD student Samuel King came up with a preliminary answer when he used a generative AI model to propose genetic blueprints for microscopic viruses. It isn’t yet an example of AI-generated life, but that could be next. Join senior AI reporter…
Share GPU clusters across teams with isolation and fairness using Amazon SageMaker HyperPod
AWS Machine Learning
A reference architecture for securely sharing one Amazon SageMaker HyperPod EKS cluster across multiple teams, using AWS IAM Identity Center for authentication, per-team SageMaker Domains and Kubernetes namespaces for isolation, HyperPod Task Governance for fairness, and namespace-level cost allocation for chargeback.
Pay-per-inference for AI agents: How BlockRun and Incarna use Amazon Bedrock AgentCore payments
AWS Machine Learning
Amazon Bedrock AgentCore payments gives AI agents a managed way to pay for services on demand, with spending limits enforced by the infrastructure. See how Incarna’s agents pay BlockRun for model inference one request at a time over x402, cutting the work of adding x402 payment support from months to days.
Señal de Hacker News
Port of the TypeScript compiler, checker and lsp to Rust, by LLM
Hacker News · 111 pts · 208 comentarios
I think I found a planet nobody knew existed. I used Claude Code to find it
Hacker News · 168 pts · 70 comentarios
AI-ready biological data: $1.8B global commitment
Hacker News · 123 pts · 18 comentarios
Vitalik Buterin backs crypto ‘bunker mode’ amid rapid AI math advances
Hacker News · 61 pts · 59 comentarios
Sub-1-Bit LLM Compression via Latent Factorization
Hacker News · 84 pts · 23 comentarios
Este snapshot se genera automáticamente desde el ranking de Discover. Es una capa de selección, no reporteo original ni análisis generado por IA.