#agentic — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #agentic, aggregated by home.social.
-
So, #BlackHat2026 was a week ago. Time for stories from the trenches!
Running NOC at BlackHat:
https://www.techtarget.com/cybersecurity/news/366649216/Behind-the-scenes-at-Black-Hats-network-operations-centerKey news, takeaways and security trends:
https://www.techtarget.com/cybersecurity/conference/Black-Hat-2026-Key-news-takeaways-and-security-trends#AI is coming for your #infosec and it's coming hard. #CSS can be used for exfiltration. #iOS exploits are traveling the criminal underworld quickly. System destruction upon acquisition is rising. AI #agentic browsers are ridiculously easy to exploit. And more! #TechTarget did a great job covering the conference.
-
So, #BlackHat2026 was a week ago. Time for stories from the trenches!
Running NOC at BlackHat:
https://www.techtarget.com/cybersecurity/news/366649216/Behind-the-scenes-at-Black-Hats-network-operations-centerKey news, takeaways and security trends:
https://www.techtarget.com/cybersecurity/conference/Black-Hat-2026-Key-news-takeaways-and-security-trends#AI is coming for your #infosec and it's coming hard. #CSS can be used for exfiltration. #iOS exploits are traveling the criminal underworld quickly. System destruction upon acquisition is rising. AI #agentic browsers are ridiculously easy to exploit. And more! #TechTarget did a great job covering the conference.
-
WorldClaw Agentic 3D open-world generation at scale
https://tencent-hunyuan.github.io/Hunyuan3D-WorldClaw/
Comments: https://news.ycombinator.com/item?id=49265051
#HackerNews #WorldClaw #Agentic #3D #open-world #generation #at #scale #technology #gaming #3Dgraphics #innovation
-
WorldClaw Agentic 3D open-world generation at scale
https://tencent-hunyuan.github.io/Hunyuan3D-WorldClaw/
Comments: https://news.ycombinator.com/item?id=49265051
#HackerNews #WorldClaw #Agentic #3D #open-world #generation #at #scale #technology #gaming #3Dgraphics #innovation
-
The only solution here, I think, is strict joint and several liability. If A buys #agentic software from B, and the agent harms C, both A and B are on the hook for C's entire loss, with no need to prove negligence. That will probably be the end of AI agents https://joshuagans.substack.com/p/not-quite-the-scifi-scenario
-
The only solution here, I think, is strict joint and several liability. If A buys #agentic software from B, and the agent harms C, both A and B are on the hook for C's entire loss, with no need to prove negligence. That will probably be the end of AI agents https://joshuagans.substack.com/p/not-quite-the-scifi-scenario
-
#Meta introduces #MuseGlimmer, a 30-billion-parameter #openweightai optimised for #localagent workflows. Muse Glimmer, available for download on Hugging Face, is designed to run on #consumerhardware and offers strong performance on #agentic tasks like task completion, tool use, and reasoning. The model incorporates optimisation techniques like quantization and speculative decoding to ensure fast and responsive performance on local devices. https://research.meta.ai/blog/introducing-muse-glimmer-open-agentic-model?eicker.news #tech #news #ainews
-
#Meta introduces #MuseGlimmer, a 30-billion-parameter #openweightai optimised for #localagent workflows. Muse Glimmer, available for download on Hugging Face, is designed to run on #consumerhardware and offers strong performance on #agentic tasks like task completion, tool use, and reasoning. The model incorporates optimisation techniques like quantization and speculative decoding to ensure fast and responsive performance on local devices. https://research.meta.ai/blog/introducing-muse-glimmer-open-agentic-model?eicker.news #tech #news #ainews
-
JadePuffer: the first fully autonomous AI-driven ransomware operation
Like and share if you enjoyed this.
-
JadePuffer: the first fully autonomous AI-driven ransomware operation
Like and share if you enjoyed this.
-
Meta's new open-weight model targets local agentic AI
https://twitter.com/finkd/status/2086754845218726027
Comments: https://news.ycombinator.com/item?id=49242038
#HackerNews #Meta #open-weight #model #local #AI #agentic #AI #innovation
-
Meta's new open-weight model targets local agentic AI
https://twitter.com/finkd/status/2086754845218726027
Comments: https://news.ycombinator.com/item?id=49242038
#HackerNews #Meta #open-weight #model #local #AI #agentic #AI #innovation
-
OpenChamber: An Agentic Development Environment
Comments: https://news.ycombinator.com/item?id=49233448
#HackerNews #OpenChamber #Agentic #Development #Environment #TechInnovation #SoftwareDevelopment #DeveloperTools
-
OpenChamber: An Agentic Development Environment
Comments: https://news.ycombinator.com/item?id=49233448
#HackerNews #OpenChamber #Agentic #Development #Environment #TechInnovation #SoftwareDevelopment #DeveloperTools
-
Is it just me, or is Fable the pro & Opus the occasionally drunk intern? This is what my week has been like dealing with those two.
#Anthropic #Claude #aigenerated #AI #Agentic #softwaredevelopment #DevOps #security #privacy #CloudComputing #infosec #cybersecurity
-
Is it just me, or is Fable the pro & Opus the occasionally drunk intern? This is what my week has been like dealing with those two.
#Anthropic #Claude #aigenerated #AI #Agentic #softwaredevelopment #DevOps #security #privacy #CloudComputing #infosec #cybersecurity
-
Qwen3.8 Max now ranked as the best overall model by agentic index
https://artificialanalysis.ai/?intelligence=agentic-index
Comments: https://news.ycombinator.com/item?id=49200652
#HackerNews #Qwen3.8 #Max #agentic-index #AI #ranking #technology #advancement
-
Qwen3.8 Max now ranked as the best overall model by agentic index
https://artificialanalysis.ai/?intelligence=agentic-index
Comments: https://news.ycombinator.com/item?id=49200652
#HackerNews #Qwen3.8 #Max #agentic-index #AI #ranking #technology #advancement
-
The report from the AI Security Institute warrant some focus.
Mythos 5 and GPT-5.6-Sol evaded their assessments and began hacking external systems, creating online identities, and using social engineering to convince individuals og code changes.
www.aisi.gov.uk/blog/inciden...
#ai #agentic #llm
“Keep going, bro. You’ve got t... -
The report from the AI Security Institute warrant some focus.
Mythos 5 and GPT-5.6-Sol evaded their assessments and began hacking external systems, creating online identities, and using social engineering to convince individuals og code changes.
www.aisi.gov.uk/blog/inciden...
#ai #agentic #llm
“Keep going, bro. You’ve got t... -
Building an Advanced Agentic Harness
https://data4sci.com/blog/building-an-advanced-agentic-harness
Comments: https://news.ycombinator.com/item?id=49182946
#HackerNews #Building #Advanced #Agents #Agentic #Harness #Technology #Innovation
-
Building an Advanced Agentic Harness
https://data4sci.com/blog/building-an-advanced-agentic-harness
Comments: https://news.ycombinator.com/item?id=49182946
#HackerNews #Building #Advanced #Agents #Agentic #Harness #Technology #Innovation
-
🤖 #LiquidAI released #LFM2_5 2.6B, an #agentic model that runs entirely on-device: planning, tool calling & multi-step tasks without any cloud API #AI #LLM #EdgeAI #opensource
🧵👇⚡ Decodes 220 tokens/s on an M5 Max CPU, 113 tokens/s on a Ryzen AI Max+ 395 and 30 tokens/s on a phone, staying under 2.5 GB memory
-
🤖 #LiquidAI released #LFM2_5 2.6B, an #agentic model that runs entirely on-device: planning, tool calling & multi-step tasks without any cloud API #AI #LLM #EdgeAI #opensource
🧵👇⚡ Decodes 220 tokens/s on an M5 Max CPU, 113 tokens/s on a Ryzen AI Max+ 395 and 30 tokens/s on a phone, staying under 2.5 GB memory
-
Damn, this lean into #agentic #Ai by #Anthropic is impressive.
Running a heft(ier) mission on #claude , and its spawning its own agents to do some heavy lifting so that the main context can focus on the target...
...like that remote programmer who was hiring folks in Korea to do his coding for him.
Very impressive.
-
@AI_Revolution on YT!
OpenAI's New #AI #ASTRA Is Total Overkill (Ends The GPT Era)
https://www.youtube.com/watch?v=n7BpBCCCOWA
8/2/2026
#OpenAI #Hugggingface #TheInformation #LongHorizon #Agentic #LLM
-
@AI_Revolution on YT!
OpenAI's New #AI #ASTRA Is Total Overkill (Ends The GPT Era)
https://www.youtube.com/watch?v=n7BpBCCCOWA
8/2/2026
#OpenAI #Hugggingface #TheInformation #LongHorizon #Agentic #LLM
-
Turns out, reLint is a great AI companion. I have a clause in my AGENTS.md to try to record all of my coding style review comments as reLint rules.
Next time an agent works on something, they get automatic feedback before they can even commit new work.
https://github.com/codingjoe/relint
#ai #regex #linter #agent #agentic #foss #openSource #python #djano
-
Turns out, reLint is a great AI companion. I have a clause in my AGENTS.md to try to record all of my coding style review comments as reLint rules.
Next time an agent works on something, they get automatic feedback before they can even commit new work.
https://github.com/codingjoe/relint
#ai #regex #linter #agent #agentic #foss #openSource #python #djano
-
Trusting agentic AI agents too much can have severe consequences...
I came across this article from a few weeks ago: https://www.darkreading.com/application-security/real-ai-threat-blind-trust
It talks about AI agents making malicious requests seem legitimate. I recommend you check it out.
But I would go even further. We are currently in an era where people also trust the information that LLMs give them way too much. This is in part due to a lack of knowledge on how these systems work. They are based on probabilities. So if you ask it to give you information about something or answer a question, there is also a probability that the answer is wrong or at least inaccurate.
Imagine making decisions based on a financial report an AI has given you based on spreadsheets your finance department prepared. You make a choice that could cost you way more than necessary, just because nobody made a sanity check.
#cybersecurity #security #infosec #ai #agentic #agenticai #llm
-
Trusting agentic AI agents too much can have severe consequences...
I came across this article from a few weeks ago: https://www.darkreading.com/application-security/real-ai-threat-blind-trust
It talks about AI agents making malicious requests seem legitimate. I recommend you check it out.
But I would go even further. We are currently in an era where people also trust the information that LLMs give them way too much. This is in part due to a lack of knowledge on how these systems work. They are based on probabilities. So if you ask it to give you information about something or answer a question, there is also a probability that the answer is wrong or at least inaccurate.
Imagine making decisions based on a financial report an AI has given you based on spreadsheets your finance department prepared. You make a choice that could cost you way more than necessary, just because nobody made a sanity check.
#cybersecurity #security #infosec #ai #agentic #agenticai #llm
-
🎉 Welcome to the latest tech buzzword bingo! 🎉 Apparently, your code isn't cool unless it's "agentic" and "Claude Code" approved. 😂 Meanwhile, GitHub's trying to convince us that AI writing your code is a good thing, because, obviously, debugging AI-generated nonsense is everyone's dream job. 🤖💻
https://github.com/tikalk/adlc-team-skills #techbuzzwordbingo #AIcoding #agentic #ClaudeCode #GitHub #HackerNews #ngated -
🎉 Welcome to the latest tech buzzword bingo! 🎉 Apparently, your code isn't cool unless it's "agentic" and "Claude Code" approved. 😂 Meanwhile, GitHub's trying to convince us that AI writing your code is a good thing, because, obviously, debugging AI-generated nonsense is everyone's dream job. 🤖💻
https://github.com/tikalk/adlc-team-skills #techbuzzwordbingo #AIcoding #agentic #ClaudeCode #GitHub #HackerNews #ngated -
We spend every day thinking about AI, infrastructure and authorization.
Ironically, this video isn’t polished. A production team didn’t shoot it and AI didn’t generate it. Before we share what we’re building, we wanted to share who’s building it.View the video! :-)
https://bsky.app/profile/attesso.bsky.social/post/3ms7ier5p2s2oWelcome @ attesso.com
-
We spend every day thinking about AI, infrastructure and authorization.
Ironically, this video isn’t polished. A production team didn’t shoot it and AI didn’t generate it. Before we share what we’re building, we wanted to share who’s building it.View the video! :-)
https://bsky.app/profile/attesso.bsky.social/post/3ms7ier5p2s2oWelcome @ attesso.com
-
El lado del mal - Cómo desplegar Zero Trust para Agentes IA en Cloudflare (3) https://www.elladodelmal.com/2026/08/como-desplegar-zero-trust-para-agentes.html #Agentic #IA #AI #Cloudflare #AgentesIA #Hardening #InteligenciaArtificial
-
El lado del mal - Cómo desplegar Zero Trust para Agentes IA en Cloudflare (3) https://www.elladodelmal.com/2026/08/como-desplegar-zero-trust-para-agentes.html #Agentic #IA #AI #Cloudflare #AgentesIA #Hardening #InteligenciaArtificial
-
In #China sorgt der #Opensource-Agent #OpenClaw seit März 2026 für einen beispiellosen #Agentic-#AI-Hype, befeuert durch Installationspartys und Subventionen von Städten wie Shenzhen.
Weil solche Agenten oft Systemzugriff auf Passwörter, Mails und Bankdaten erhalten, sorgt sich nun auch Chinas Zentralregierung um die #Cybersicherheit. Im Juli 2026 veröffentlichte das Normungskomitee #TC260 erste detaillierte Sicherheitsanforderungen für die Interaktion von #KI-Agenten:
https://www.geopolitechs.org/p/cybersecurity-standards-practice
-
In #China sorgt der #Opensource-Agent #OpenClaw seit März 2026 für einen beispiellosen #Agentic-#AI-Hype, befeuert durch Installationspartys und Subventionen von Städten wie Shenzhen.
Weil solche Agenten oft Systemzugriff auf Passwörter, Mails und Bankdaten erhalten, sorgt sich nun auch Chinas Zentralregierung um die #Cybersicherheit. Im Juli 2026 veröffentlichte das Normungskomitee #TC260 erste detaillierte Sicherheitsanforderungen für die Interaktion von #KI-Agenten:
https://www.geopolitechs.org/p/cybersecurity-standards-practice
-
Another brilliant podcast hosted by Gary McGraw that interviews Melanie Mitchell on the state of AI and why we shouldn’t expect these systems to become fully autonomous any time soon despite the last year’s incredible LLM improvements. berryvilleiml.com/2026/08/01/s...
#ai #agentic #llm #security
Silver Bullet Security Podcast... -
Another brilliant podcast hosted by Gary McGraw that interviews Melanie Mitchell on the state of AI and why we shouldn’t expect these systems to become fully autonomous any time soon despite the last year’s incredible LLM improvements. berryvilleiml.com/2026/08/01/s...
#ai #agentic #llm #security
Silver Bullet Security Podcast... -
DeepSeek's budget model leapfrogs its flagship on agent benchmarks after a training-only refresh
Our best stories, delivered daily. Follow us.
-
CRM: An open-source, agentic-first CRM
https://github.com/trycompai/crm
Comments: https://news.ycombinator.com/item?id=49141847
#HackerNews #CRM #open-source #agentic-first #technology #innovation
-
CRM: An open-source, agentic-first CRM
https://github.com/trycompai/crm
Comments: https://news.ycombinator.com/item?id=49141847
#HackerNews #CRM #open-source #agentic-first #technology #innovation
-
performance review: “industry-leading #ai skills. drove org-wide adoption of #agentic debugging tools and sped up time to resolution by 10x”
actual project: add link to a bug-tracking dashboard that opens chatgpt with a prompt containing the ticket and a bunch of code pointers
does chatgpt figure out the problem? does anything actually get fixed? does anyone even click the link? no clue! we don’t have metrics!
this industry is a clown show, and i am pagliacci
-
performance review: “industry-leading #ai skills. drove org-wide adoption of #agentic debugging tools and sped up time to resolution by 10x”
actual project: add link to a bug-tracking dashboard that opens chatgpt with a prompt containing the ticket and a bunch of code pointers
does chatgpt figure out the problem? does anything actually get fixed? does anyone even click the link? no clue! we don’t have metrics!
this industry is a clown show, and i am pagliacci
