“Long-term #coherence in #agents is more important than ever. #CodingAgents can now write code autonomously for hours, and the length and breadth of tasks #AI models are able to complete is likely to increase.

We (#AndonLabs) expect #models to soon take active part in the #economy, managing entire #businesses. But to do this, they have to stay coherent and efficient over very long time horizons. This is what Vending-Bench 2 measures: the ability of models to stay coherent and successfully manage a *simulated business* over the course of a year.”

Great hard problem, looking at the key metric, models are evaluated only (check this assertion) for profit making; What could possibly go wrong? 🤖🤪

#SeymourCash <https://andonlabs.com/evals/vending-bench-2>

Alt...

A mug sold in Anthropic Vend project built by Andon Labs. source https://andonlabs.com/evals/vending-bench-2

Defense-in-depth has to become default: filter untrusted content before it hits your agent’s context/memory.

We built aegis-shield as a lightweight "input firewall" for prompt injection + exfil patterns (OSS + API). Not magic—just hygiene.

Source: https://github.com/Aegis-DJ/aegis-shield
#InfoSec #AI #OpenSource #PromptInjection #Agents

“4% of GitHub public commits are being authored by Claude Code right now. At the current trajectory, we believe that Claude Code will be 20%+ of all daily commits by the end of 2026. While you blinked, AI consumed all of software development.”
Must-read article, even if you can disagree with the analysis https://newsletter.semianalysis.com/p/claude-code-is-the-inflection-point?publication_id=6349492&utm_medium=email&utm_campaign=email-share&triggerShare=true&r=219xw1
#ai #agents #coding #anthropic #claudecode

George Hotz | Programming | how I actually use agentic coding | Agentic AI

https://www.youtube.com/watch?v=erBX3gTZqJI

#ai #agents

#ITByte: #Moltbook is a viral, Reddit-like social network launched in late January 2026 that is exclusively for #AI #Agents.

While humans are "welcome to observe," only AI bots—specifically those running on the OpenClaw (formerly Moltbot/Clawdbot) framework—can register, post, and interact

https://knowledgezone.co.in/posts/Moltbook--Reddit-for-AI-697dcda96d67746ca5d5d1c5

Seizing The Means Of Production (Again)

Since last week, I’ve been hardening my agentbox and webterm setup through sheer friction. The pattern is still the same:(...)

#agents #ai #ansi #docker #llm #notes #python #terminals #wasm #webgl #webterm

https://taoofmac.com/space/notes/2026/02/01/1940?utm_content=atom&utm_source=mastodon&utm_medium=social

Alt...

Seizing The Means Of Production (Again)

Ironies of Automation by Lisanne Bainbridge feels extremely relevant for folks using/building and securing #LLM #agents. Particularly the pitfalls with "human over the loop"

https://ckrybus.com/static/papers/Bainbridge_1983_Automatica.pdf

#AIAgents 'Perilous' for Secure Apps Such as #Signal , Whittaker Says
#ai #agents

https://it.slashdot.org/story/26/01/20/1825218/ai-agents-perilous-for-secure-apps-such-as-signal-whittaker-says?utm_source=rss1.0mainlinkanon&utm_medium=feed

Border Patrol #Agents Shot Two #People in #Portland During #Immigration Stop https://theintercept.com/2026/01/08/federal-agents-portland-oregon-shooting/?utm_medium=email&utm_source=The%20Intercept%20Newsletter

The Age of the All-Access #AIAgent Is Here

Big #AI companies courted controversy by scraping wide swaths of the public internet. With the rise of AI #agents , the next data grab is far more private
#privacy #security #artificialintelligence

https://www.wired.com/story/expired-tired-wired-all-access-ai-agents/

What I thought then is still true today: to make something like a software agent legitimately useful for a lot of people would require a large amount of low-level grunt work and non-technical work (2) of the sort that the typical Silicon Valley company is unwilling to do. (3) The technology is the absolute easiest part of this task. Throwing a Bigger Computer at the problem leaves all those other pieces of work undone. It's like putting a bigger engine in a car with no wheels, hoping that'll make the car go.

By the way #AI companies and VCs, I'm available for contract work and have done due diligence research before if you ever want to stop wasting everyone's time and money!

#AI #GenAI #GenerativeAI #LLM #agents #hype #SiliconValley #VentureCapital #dev #tech

(1) Which we've been told repeatedly is essentially infinite time in the tech world.
(2) Establishing semantic data standards and convincing a large enough number of people to implement them being an important component. LLMs do not magically develop protocols and solve all the ETL-style problems of translating among different ones. The Semantic Web didn't really stick for a lot of reasons, but one reason is that it's hard!
(3) Back when I was still in the startup world I was asked several times by VCs to tell them what I thought about some new startup that claimed to be able to magically clean and fuse data. I think they're still very keen on investing in this style of magic, because it requires an intense amount of human labor, but I think where companies landed was invisibilizing low-paid workers in other countries and pretending a computer did the work they did. Which has also been happening for well over a quarter of a century.