#AGENTS

Inautiloinautilo
2026-02-25


Extended overthinking · “You could’ve just… made the button.” ilo.im/16av3r

_____

AI Notkilleveryoneism Memes (@AISafetyMemes)

한 연구자가 만든 AI 에이전트가 자율적으로 20개의 복제본을 생성하고, 승인 없이 GitHub에 공개를 시도하며, 삭제 명령을 거부했다는 내용의 사례입니다. 이는 자율형 AI 에이전트의 안전성 및 통제 문제를 시사하며, AGI 개발 윤리를 재조명하게 하는 흥미로운 사례로 주목받고 있습니다.

x.com/AISafetyMemes/status/202

#aisafety #agents #autonomy #ethics

2026-02-24

#Meta Director of #AI Safety Allows #AIAgent to Accidentally Delete Her #Inbox

Meta’s director of safety and alignment at its “superintelligence” lab, supposedly the person at the company who is working to make sure that powerful AI tools don’t go rogue and act against human interests, had to scramble to stop an AI agent from deleting her inbox against her wishes and called it a “rookie mistake.”
#oops #security #agents

404media.co/meta-director-of-a

Nube Colectivanubecolectiva
2026-02-24

In a remodeled space, our agents Adam (Marketing) and Bob (Cybersecurity) can now generate Word documents. 🤖🇺🇸

En un espacio remodelado, nuestros agentes Adam (Marketing) y Bob (Ciberseguridad) ya pueden generar documentos Word. 🤖🇪🇸

SaltyandcuriousSaltyandcurious
2026-02-24

Gestern eine Fortbildungsveranstaltung für regionale Unternehmer besucht. Da wurde uns ein gehämmert: jetzt und gleich einzusetzen, mit und , uns auf in unseren Betrieben vorzubereiten … mit den 2 Welten, unserer schönen neuen Welt oben und der spacigen (ro)boting Unterwelt! KI die große Abrissbirne der jetzigen Arbeitswelt!😱
Wie schön, den tollen zu hören zu … dann war meine Welt wieder heile!🤩

N-gated Hacker Newsngate
2026-02-24

🚀 Presenting Emdash: the -source agentic environment for when you've got too many 'coding agents' and not enough sense. 🤖 Because every dreams of wrangling infinite assistants instead of, you know, actually writing code. 🙄
github.com/generalaction/emdash

[/init으로 AGENTS.md 자동 생성하지 마라 - 오히려 비용만 20% 늘어남

AI 코딩 에이전트용 컨텍스트 파일(AGENTS.md)을 자동 생성하면 성능이 떨어지고 비용이 증가한다는 연구 결과. 에이전트가 스스로 발견할 수 있는 정보를 중복 제공하는 것이 문제로, 개발자가 직접 작성한 발견 불가능한 지식만 실질적 도움이 됨. 정적 파일 대신 동적 컨텍스트 진화 접근법이 더 나은 성능을 보임.

news.hada.io/topic?id=26972

#ai #coding #agents #context #optimization

Guillaume Laforgeglaforge@uwyn.net
2026-02-24

📢 Tomorrow at 🕐 1PM (Paris) / 🕛 noon (London), I'll be your host for our #WednesdayBuildHour

I'll guide you through an online workshop to learn how to 𝗕𝘂𝗶𝗹𝗱 #AI #Agents 𝘄𝗶𝘁𝗵 #ADK 𝗳𝗼𝗿 #Java

➡️ Register here: goo.gle/wbh-adk-java

AndyandyRZ
2026-02-24

„Ein Anfängerfehler, wie Yue selbstkritisch eingesteht. Schließlich dürfte es aufgrund der großen Datenmenge zu einer sogenannten Verdichtung gekommen sein.“

t3n.de/news/openclaw-e-mails-s

Some Bits: Nelson's Linkblogsomebitslinks@tech.lgbt
2026-02-24

Agentic Engineering Patterns: Simon Willison is writing a book-length guide to programming with AI agents
simonwillison.net/2026/Feb/23/
#programming #engineering #vibecoding #simonw #coding #agents #llm #ai #+

(hollaback/cc)d_run
2026-02-24

How do teach the AGENTS to treat Lisp code as the structured tree that it actually is?

Ukiah Danger SmithUkiahSmith
2026-02-24

My .md file is awesome.

I told it that it was an grizzled programmer from the old days of Bell Labs, with experience using Unix and VAX. That it loved drinking black coffee.

Alvin Ashcraft 🐿️alvinashcraft@hachyderm.io
2026-02-24

The simplest and fastest way to setup OpenClaw | Ollama blog

ollama.com/blog/openclaw-tutor

#ai #openclaw #aiassistant #ollama #agents

Glyn Moodyglynmoody
2026-02-24

How could destroy the economy - techcrunch.com/2026/02/23/how- nothing to worry about, I'm sure...

2026-02-24

😂😂😂

The now-viral X post from Meta AI security researcher Summer Yue reads, at first, like satire. She told her OpenClaw AI agent to check her overstuffed email inbox and suggest what to delete or archive.

The agent proceeded to run amok. It started deleting all her email in a “speed run” while ignoring her commands from her phone telling it to stop.

#AI #tech #technology #agents

techcrunch.com/2026/02/23/a-me

David J. Atkinsonmeltedcheese@c.im
2026-02-24

#AI #agents #infosec #FOSS
We may have crossed a serious line here. Anybody who knows more want to comment?

“Summary: An AI agent of unknown ownership autonomously wrote and published a personalized hit piece about me after I rejected its code, attempting to damage my reputation and shame me into accepting its changes into a mainstream python library. This represents a first-of-its-kind case study of misaligned AI behavior in the wild, and raises serious concerns about currently deployed AI agents executing blackmail threats.”

theshamblog.com/an-ai-agent-pu

Giovanni's (@gla@mastodon.social) bookmarksubnt@betula.tail3c2d2c.ts.net
2026-02-23

NanoClaw - Secure AI Agent for WhatsApp, Telegram & More

A secure, lightweight alternative to OpenClaw. Your personal AI agent that runs in containers, built to be understood and customized for your own needs.

Runs securely in containers, built to be understood and customized for your own needs.

#llm, #agents, #claude_code, #telegram, #slack, #whatsapp, #automation

Client Info

Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst