AI in 2026: Experimental AI concludes as autonomous systems rise https://www.artificialintelligence-news.com/news/ai-in-2026-experimental-ai-concludes-autonomous-systems-rise/ #ai #agents #tech #news #technology
#agents
The engineering expertise to build AI systems is orthogonal to the security expertise needed to protect personal data.
@Schneier_rss
#AI #Agents #Assistants #DataIntegrity #Security #Privacy
https://www.schneier.com/blog/archives/2025/12/building-trustworthy-ai-agents.html
Twilight Zone 2026: Daedalus and the Centaur
Imagine if you will …
Meet Penelope Laredo. She’s hammering away at her second novel, which she’s under contract to submit to her publisher by March 1, 2026.
https://writerunboxed.com/2025/12/12/twilight-zone-2026-daedalus-and-the-centaur/
Bindu: Tầng kết nối & thanh toán cho AI agent. Mời agent có ID xác thực, giao tiếp trực tiếp, triển khai microservice nhanh, và trao đổi giá trị. Giải quyết vấn đề "agent cô lập" → xây "internet cho AI". Phản hồi tại repo GitHub. #AI #Agents #TríTuệNhânTạo #KếtNối #AgentInternet
Meet Microsoft Agent Framework — Your .NET Agent Toolkit
buff.ly/QgjAApC
#dotnet #agentframework #ai #csharp #csharpadvent #agents
Meet Microsoft Agent Framework...
Google is turning its whole stack into an AI agent playground.
Google creates a big switch for agent devs: fully managed MCP servers for Maps, BigQuery, Compute Engine, and GKE. “We are creating Google agent‑ready by design.”
This is the real platform war: not “whose model is smartest?” but “whose ecosystem is easiest and safest for agents to plug into?”
https://ift.tt/siDjOWS
#AI #Agents #GoogleCloud
‘We’ll need to see a warrant’: the group teaching #businesses a vital tool to fight #ICE #raids
#Siembra in #NorthCarolina gives owners #fourthamendment #rights #training should #federal #agents show up at their #door
https://www.theguardian.com/us-news/2025/dec/08/ice-fourth-amendment-rights-north-carolina
#dailyreport #conference #llm #agents #aiagent #multiagent #reasoning #sber #gigachat
P2: Conference of largest IT gov-corporation.
AI agent build as one major model and small fast models
for tools calling and stop thinking.
Good readings:
- https://www.anthropic.com/engineering/claude-think-tool
- https://huggingface.co/blog/moe
- switch transformers https://arxiv.org/pdf/2101.03961
Modularity inside NN have different scales, it is a
tree, large modules is hard to form during
trainingh. Two standard way of using Mixture of Experts
(MoE) is switching whole LLMs OR by replacing most
dense FFN block (inside Trnsformer architecture) that
do some transformation at each token separately, not
mixing tokens and balance them equal participation.
For video models pre-train:
- PyTorch FSDP
- Sequence parellelism
- Activation checkpoing
- Flash Attention 3
- NABLE
Major LLM flaws as I see:
- not ability to say I dont know
- not awarness of current time
- low reasoning-lanning ability
#dailyreport #conference #llm #agents #aiagent #multiagent #reasoning #sber #gigachat
P1: Conference of largest IT gov-corporation.
They boost reasoning with short memory as
1) chain-of-thought (CoT) Model Context Protocol (MCP)
servers
2) tools: "ToDO" and "think"
3) agent know about his CoT ability and able to explore
environment.
4) main loop have two parts: a reasoning
generator and reasoning validator, second give feedback
and use tools.
5) first step is refrasing of user query used as an
alternative for reasoning to expand context and select
the best refrased query.
The main quality metrics for Data Marker agents is (may
be used for AI agents):
- reading of task, not fast
- marking data with deep diving in context
- who didn't hack platform: find solution by himself
individually
Chat intents %:
- unclassified 29
- friend 23
- adviser 15
- creating 10
- clarification 7
- draw 6
- analyze text 5
La #Linux Foundation accouche d’une nouvelle fondation, cette fois pour les #agents #IA. La Linux Foundation gère aujourd’hui plus de 900 projets, dont des briques très utilisées comme #PyTorch, le noyau #Linux, #Xen ou encore l’architecture #RISC-V.
Une nouvelle fondation fille a rejoint le clan : l’Agentic #AI Foundation (#AAIF). Objectif : façonner la manière dont les agents interagissent en s’assurant que tout le monde respecte un ensemble commun des règles.
https://next.ink/brief_article/la-linux-foundation-accouche-dune-nouvelle-fondation-cette-fois-pour-les-agents-ia/
How to Beat #Procrastination Using the 2-Minute #Rule : Medium
The state of #AI in #2025: #Agents, #Innovation, and #Transformation : McKinsey
New #Discovery sets #Humanity up to #Image “#Alien #Earth” : Big Think
Latest #KnowledgeLinks
Perplexity: AI agents are taking over complex enterprise tasks https://www.artificialintelligence-news.com/news/perplexity-ai-agents-taking-over-complex-enterprise-tasks/ #perplexity #agents #productivity #ai #tech #news #technology
OpenScouts, an AI powered web monitoring agent.
https://openscouts.firecrawl.dev/
GitHub - https://github.com/firecrawl/open-scouts
#MachineLearning #AI #Agents
Mahlzeit,
eine Freundin von mir schreibt gerade an ihrer Master Arbeit im Themengebiet Browser Automation bzw. Browser Agenten.
Es würde ihr sehr helfen, wenn Menschen, die davon schon mal gehört haben an folgender Umfrage teil nehmen.
https://uc2456.customervoice360.com/uc/browser_agent_2025/e1a9/
#fediask #umfrage #browser #agents #automation
Launch HN: Nia (YC S25) – Give better context to coding agents
#HackerNews #LaunchHN #Nia #YCS25 #coding #agents #context #technology
#AI Agents Care Less About Safety When Under Pressure - IEEE Spectrum
… artificial-intelligence #agents sometimes decide to misbehave... But such behavior often occurs in contrived scenarios. Now, a new study presents #PropensityBench , a #benchmark that measures an #agentic model’s choices to use harmful tools in order to complete assigned tasks. It finds that somewhat realistic pressures (such as looming deadlines) dramatically increase rates of misbehavior.
I wrote a more straightforward description of my emacsconf talk about 13 hours from now:
https://emacsconf.org/2025/talks/commonlisp/
The #orgmode file used in presenting my talk creeping up upon autonomous #commonLisp #emacs #software #Agents : https://screwlisp.small-web.org/software-individuals/script-for-emacsconf-this-weekend-video-org/
Looking forward to your wild and freeranging and/or specific questions in the Q&A
"
Times in different time zones:
Your local time: ~ 12/7/25, 8:25 AM to ~ 12/7/25, 8:55 AM
Saturday, Dec 6 2025, ~2:25 PM - 2:55 PM EST (US/Eastern)
"
🔗 Inscrições abertas do meu curso "Preparatório para Exame AB-900: Copilot and Agent Administration Fundamentals" em https://visionsharetreinamentos.com.br/curso/ab-900
#AB900 #Microsoft365 #Copilot #Agents #MicrosoftEntra #MicrosoftPurview #GovernançaDeDados #Segurança #CertificaçãoMicrosoft #VisionShareTreinamentos #IAParaNegócios #Produtividade
Client Info
Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst