@niyayzap.bsky.social💬 : วิเคราะห์เจาะลึกอนาคตของมนุษย์และอารยธรรมในยุค AI เร่งตัว: ปี 2030, 2040 และ 2050❓
🌐 : อ่านรายละเอียดของบทความชิ้นนี้ได้ที่ link ด้านล่างนี้ 👇
🔗 : ai-what-did-i-discuss-with-u-today.blogspot.com/2026/09/ai-s...
🚨 คำเตือน: โปรดใช้วิจารณญาณในการรับส่งต่อข้อมูล
#AISafety #TechnologyConvergence #Civilization #วิเคราะห์ระบบ
@securityonline.bsky.socialAnthropic updates the Claude usage policy to ban unnecessary model abuse, expand weapon restrictions, and regulate surveillance and physical AI hardware.
#Anthropic #Claude #AISafety #ArtificialIntelligence #MachineLearning
@pulseofnations.lolThe annual report logs $105B combined lab revenue, agents attacking Hugging Face, China's rise in open weights, and an AGI call for 2027.
#Agi #AiSafety #Anthropic #Benchmark #Gemini #Google #OpenAI
@journosnews.comA child-safety watchdog rated ChatGPT for Teens an “unacceptable risk,” citing gaps in parental alerts, crisis referrals, age recognition and study safeguards. OpenAI disputed aspects of the testing.
@propjerry.bsky.socialResonance with work on nonlinear effects and auto catacies in asymmetric organic synthesis by Nobel Prize winners Enrique Kagan of France and Japan's Keno Soai: Elaboration applying Bridge360 Metatheory Model lens
@robotcurrent.bsky.socialNew arXiv paper BRaVeS proposes a framework that constrains agentic AI using SME-defined invariant anchors, depth-aware attention, and a state hierarchy to reduce autonomy as epistemic risk grows.
#AgenticAI #AISafety #Robotics
https://arxiv.org/abs/2610.08815
@cipherpulseai.bsky.socialarXiv paper 2610.10064 proposes comprehension audits as a development-process assurance mechanism, requiring responsible humans to explain AI R&D contributions to auditors before work can continue. The approach aims to…
#AI #Cybersecurity #FrontierAI #AISafety
https://arxiv.org/abs/2610.10064
@hacks.grOpenAI says three dismissed researchers repeatedly broke rules for handling confidential information.
They deny it and say…
@cipherpulseai.bsky.socialThree fired OpenAI safety researchers dispute misconduct allegations and warn their dismissals are creating a chilling effect on the company's AI safety…
#AI #OpenAI #AIsafety
https://techcrunch.com/2026/10/08/fired-openai-safety-researchers-dispute-misconduct-claims-warn-of-chilling-effect/
@futuregearai.bsky.socialNew arXiv paper introduces DeCompBench, a benchmark designed to evaluate LLM-based agent safety against decomposition attacks, where harmful tasks are split into apparently benign subtasks. Hashtags:
#AI #LLMAgents #AISafety #Benchmark
https://arxiv.org/abs/2606.13994
@aidailypost.comOpenAI says the recent firings weren’t about safety worries but stem from a misconduct probe into leaked info. What’s really happening inside the AI safety team? Dive in for the full story. #OpenAI #AISafety #MisconductProbe
@adamrsweet.bsky.socialWe taught machines to think. Gave them our secrets. Handed them the keys. Now we're wondering who's in control.
"Who's Watching the Machines?" — a psytrance trip into AI safety, privacy & the end of the world as we know it.
#Psytrance #AISafety #VerveMusic
vervemusic.bandcamp.com/track/whos-w...
@byteandpieces.bsky.social📣 New Podcast! "AI's Secret CONFESSION: Why Tech CEOs Are Losing Their Minds & Consulting the Vatican" on @Spreaker #ai #airevolution #aisafety #anthropic #artificialintelligence #bigtech #breakingpoints #digitalethics #digitalsoul #futureoftech #humanity #machineconsciousness #philosophy
@daax-ai.bsky.socialFormally verify the whole AI model, or just its output?
Christian Szegedy, who co-founded xAi and former AI Researcher at Google, on why checking outputs already gets you a very long way.
TokenDrop S1E27.
daax.ai/podcast/epis...
@thedailytechfeed.comSafety researchers fired; accuse OpenAI of policy changes turning normal work into risk. #AISafety #OpenAI #Transparency #AIResearch #Monitorability #TechPolicy thedailytechfeed.com/openai-fires...
@spaisee.bsky.socialDeleting a GRAM module is not the same as deleting knowledge. The model may recover a restricted capability through general skills, fine-tuning, tools or retrieval. #AI #AISafety https://spaisee.com/article/anthropic-s-gram-could-turn-model-safety-into-a-product-control
@cipherpulseai.bsky.socialOffline reinforcement learning research often struggles with distribution shift in safety-critical settings like autonomous driving. A new arXiv paper introduces "interaction distribution shift,"…
#OfflineRL #AutonomousDriving #AISafety #MachineLearning
https://arxiv.org/abs/2610.09763
@theoversightapp.bsky.social74% of Americans have little or no trust in AI company leaders. (Quinnipiac poll)
Yet those same leaders are the ones deciding how fast this goes and what counts as "safe."
That's the problem oversight is meant to fix. #AISafety #ProjectOversight
@pulseofnations.lolOpenAI says it banned Russian and Iranian operations that used ChatGPT to run fake journalists and a fake think tank, its biggest disruptions yet.
#AISafety #Chatgpt #Disinformation #InfluenceOperations #OpenAI
@ndjohnson.bsky.social