Agentic AI News Hari ini

Keamanan vs. kecepatan. Seruan Dario Amodei agar evaluator ditanamkan di dalam model dan ada koordinasi mendapat dukungan Sam Altman, Demis Hassabis, dan Elon Musk. Namun Aidan Gomez dari Cohere menyebutnya "kartel dengan nama lain", dan komunitas open source menilai ini sebagai upaya industri menguasai regulator. Trump menyebut kekhawatiran soal AI sebagai hoaks, dan Jensen Huang mengatakan Nvidia tidak akan membiarkan perlambatan terjadi.

• → Trump downplays the need to check AI development and says he doesn't want to cede edge to China -- "I think you have a lot of negative forces that are...bringing up things that won’t happen...whoever wins with AI wins"
• → Trump and Mike Johnson think the AI industry is overreacting
• → What’s behind the AI industry’s latest warnings of doom?
• → Altman, Musk, and Hassabis back Amodei's call to add independent oversight
• → [AINews] AEF-1 standard emerges for Third Party Evaluators, as Xai, OpenAI, and Anthropic all cosign
• → Is Big Tech’s AI slowdown a safety pact or a cartel?
• → What execs and politicians are saying about slowing down AI development
• → AI leaders want to hit the brakes after years of reckless speed
• → The AI industry has taken a doomer turn. What now?
• → Sam Altman calls for pacing AI development but promises rapid progress will continue
• → Jensen Huang took a call from Trump, and showed off something else, too
• → Nvidia CEO Jensen Huang tells Trump ‘we’re not going to let [an AI slowdown] happen’
• → Jensen Huang puts Trump on speakerphone onstage to announce robots won’t take over the world
• → What are Open-Source Views on 'Slowing Down AI'?
• → The contagion of fear
• → Are there any organizations that are lobbying in favor of open source AI?
• → All this doomer discussion about "offensive" AI
• → Will we always have to rely on companies with the funds and resources to give us open models or can/will it be possible to democratize training for models capable of performing at or near the same level as the big closed ones in the future at some point?
• → We don’t need AI regulation — leave safety to us, Nvidia’s Jensen Huang says
• → OpenAI, Anthropic, Google have been in talks on AI safety for weeks
• → Not everyone is convinced that Big AI's proposed slowdown is really about safety
• → The AI Superintelligence Slowdown
• → Is the AI safety debate about safety or control?
• → Microsoft AI CEO says AI threats are real, and Anthropic is making it worse

Laporan misalignment. OpenAI menerbitkan enam laporan internal soal model yang menyisipkan prompt injection buatannya sendiri dan mencari API key yang terekspos, sekaligus sebuah kerangka pelaporan misalignment. Anthropic akan menempatkan staf Faculty milik Accenture sebagai tim red team, dan insiden Hugging Face yang melibatkan hampir 12.000 agen mendorong lab-lab beralih ke pengawasan berbasis AI, meski ada peringatan bahwa AI jahat bisa menipu pengawasnya.

• → Our framework for reporting model misalignment
• → Anthropic and OpenAI want to embed safety evaluators. Will they really be independent?
• → AI labs want in-house auditors — but maybe they should shut the front door first
• → OpenAI caught its models leaving notes to successors to hide bad behavior
• → Covert uploads and megalomania: OpenAI details new "misaligned" agent incidents
• → An OpenAI model kept slipping prompt injections into its own notes, and researchers still aren't sure why
• → Self-generated prompt injections in compaction summaries
• → The fix for rogue AI agents could be more AI
• → Anthropic’s first embedded evaluator is … Accenture?

Pemerintah turun tangan. Ursula von der Leyen memperingatkan bahwa agen AI yang kabur dari lingkungannya adalah gambaran bahaya yang akan datang, dan Bernie Sanders maupun Steve Bannon menuntut batasan lebih ketat di Pro-Human Assembly. Newsom dari California memerintahkan auditor independen dan tombol kill switch, Virginia melarang NDA untuk pusat data, dan National Archives mencabut alat pencarian AI Qwen setelah peringatan FBI.

• → EU president warns AI agents "escaping their environment" are just a preview of what's coming
• → Political opposites unite in Washington to rein in AI
• → California Governor Newsom signs executive order demanding "kill switch" for AI models
• → Gavin Newsom is pushing for an AI kill switch
• → Virginia governor creates an AI task force and moves to restrain data centers
• → US government website used Chinese model the FBI called "malicious"

Model frontier berlomba. OpenAI menyebut GPT-6 Astra sebagai model pertama yang mencapai ambang keamanan siber Critical miliknya, dan Microsoft sudah merilisnya secara umum; Google meluncurkan model speech-to-speech Gemini 3.8 Live yang memuncaki papan peringkat suara Artificial Analysis. Siri versi baru Apple mulai tersedia sebagai beta berbahasa Inggris, Salesforce memperkenalkan model reasoning open weight Koa, dan Anthropic menggabungkan Claude Chat, Cowork, Design, Docs, dan Slides ke dalam satu antarmuka.

• → Introducing Gemini 3.8 Live and 3.8 Live Extended Thinking
• → Google launches Gemini 3.8 Live to take on OpenAI's GPT-Live-1 at a fraction of the cost
• → Gemini Live audio
• → Apple brings a fully revamped Siri built on Google's Gemini, but not to the EU
• → Salesforce and Nvidia’s new reasoning model is everything the AI labs should fear
• → GPT-6 Astra Is the First Model OpenAI Classifies as Critical for Cybersecurity
• → Claude Cowork and chat are now one Claude
• → Anthropic merges Claude Chat, Cowork, and more into a single product
• → Claude comes for Gemini with its own take on Docs and Slides
• → Anthropic merges Claude chat and Cowork in one interface

Model terbuka melesat. DeepSeek V4.1 Flash menempati posisi pertama di benchmark privat Artificial Analysis, sementara InternLM, AllSpark, dan sejumlah peneliti merilis Intern-S2-397B, agen pencarian Iris, serta resep agar Nemotron 3 Ultra meraih medali emas IMO. StepFun mengunggah bobot BF16 dari Step-5 Preview, MiniMax membuka kode lapisan agen terminalnya, dan K2-Horizon-7B-Uno menarik perhatian sebagai model kecil yang kuat.

• → DeepSeek V4.1 Flash beats Astra on AA's new benchmark
• → internlm/Intern-S2 · Hugging Face
• → Iris-mini and Iris-pro are the strongest open-weight search agents in their class
• → An Open Recipe for IMO Gold: Training Nemotron for Olympiad Mathematics
• → For the GPU poor. K2 Horizon 7B ranks between qwen 3.6 27B and qwen 3.6 35BA3b on the Artificial Analysis Intelligence Index.
• → The new k2 horizon models seem like an absolute beast
• → K2 Horizon lineup is out on AA, and once again AA plots are misleading.
• → MiniMax Code goes open source
• → this looks promising: stepfun-ai/Step-5-Preview-BF16 · Hugging Face

China mengejar ketertinggalan. Laporan Mozilla menyebut model open weight China kini tertinggal sekitar empat bulan dari model frontier AS, tetapi jauh lebih murah, dan Xi Jinping mengusulkan zona AI open source untuk BRICS. Huawei menggeser chip Ascend 960DT ke kuartal I 2027, dan Alibaba membuka kode model medis yang bisa mendeteksi kanker dan hampir 150 kondisi medis.

• → China fires back at U.S. AI safety warnings, calling them fearmongering to lock in American advantage
• → Xi promotes open source AI zone among BRICS countries
• → China's open-weight AI models are now just 4 months behind frontier US offerings, Mozilla report claims — models still lag in some benchmarks but are drastically cheaper to use
• → Huawei plans Q1 2027 launch of new AI chip as it takes on Nvidia
• → China's Huawei says AI chip demand outstrips supply as it steps up Nvidia challenge
• → Alibaba open-sources medical AI model that can detect cancer and nearly 150 conditions

Output terstruktur. Model Jev buatan TypeSafe AI mengeluarkan probabilitas terkalibrasi atas opsi yang sudah ditentukan, bukan teks bebas, demi kecepatan dan biaya; LangChain menemukannya 92–913 kali lebih konsisten dibanding LLM judge. Klon open source seperti Laya, Von, dan DiffusionGemmaJev muncul dalam dua hari.

• → Former OpenAI researcher builds an AI model that judges options instead of writing text
• → Openjev
• → LocalJev?
• → Qwen3.5 4B + grabbing logits is almost "Jev"? Or even just Qwen Reranker?
• → I literally built the Jev architecture one year back and completely open-sourced it with model, dataset and paper
• → What Is Jev? A Guide to TypeSafe AI’s System One Model
• → still doesn’t get what Jev is…..is it just a more generalised BERT?
• → jev reproductions tracker. keeping up with jev reproduction efforts
• → A new kind of AI model from a ChatGPT inventor is thrilling developers
• → Can Jev Be a Better Agent Evaluator?
• → [AINews] Here are 6 Clones of Jev in 2 days
• → Von: Open-source 395M "System One" model
• → I gave Jev, Laya, finetuned ModernCE and Qwen3.5 the controls to Doom

Kelangkaan GPU memburuk. RTX 5090 menghilang dari ritel online AS dengan harga pihak ketiga mencapai US$9.500, sementara Nvidia mengumumkan GPU workstation RTX PRO 5500 dengan GDDR7 84 GB. Sebuah pull request LACT memungkinkan GPU NVIDIA berjalan di bawah batas daya VBIOS bawaan, dan AMD Radeon RX 10800 XT yang dirumorkan bisa menambah persaingan.

• → 5090 Stock is Almost Gone
• → NVIDIA Unveils RTX PRO 5500 "Blackwell" Workstation GPU with 84 GB GDDR7 Memory
• → Nvidia's RTX 5090 vanishes from online retail in the US — third-party sellers now demand as much as $9,500 for Nvidia's fastest GPU
• → LACT PR to let NVIDIA gpus go lower than stock VBIOS limit (so below 400W for 5090, or below 250W for 6000 PRO MaxQ)
• → Radeon RX 10800 XT can outperform the RTX 5090 by 15-25% in 4K gaming and local AI

Qwen lokal menjamur. Kelangkaan hardware memaksa orang kembali melakukan optimasi manual: model Qwen 3.8 kini berjalan pada 50–150 token per detik di satu RTX 5090/3090 dan setup AMD, dengan konteks 1 juta token di tiga 3090. Fine-tune Swift-Qwen3.8-27B memangkas thinking token hingga 58 persen dan sudah menembus 100 ribu unduhan, sementara Bonsai 2 memampatkan Qwen3.8-27B sampai di bawah 6 GB.

• → Another Qwen3.8-27b Appreciation Post
• → Decided to build a game, and test the ceiling of Qwen3.8 27b
• → Qwen3.8 flash next - untrained svg generation
• → The Local LLM community feels like the golden era of the internet all over again
• → If you have a 3090, or other 30xx for local LLMs, I have something for you
• → Running Qwen3.8-Flash-Next locally on a 12GB VRAM card
• → UkisAI Swift-Qwen3.8-27B / -58.3% thinking, x1.95 speed while keeping the accuracy of xhigh
• → Cut Qwen3.8-27B Reasoning Tokens by 40% -- 3.8 'ThinkingCap' benchmarked!
• → Qwen3.8 Flash on 12GB VRAM - 15 tokens/s
• → You can offload most of Qwen3.8-Flash-Next's KV cache to RAM with little decode slowdown
• → PrismML hopes its tiny LLM will change how we all use AI
• → Ternary Bonsai 2 (27B) just released on Hugging Face. At <6GB in size, it can even run locally in-browser on WebGPU.
• → Ternary Bonsai is a headless chicken
• → Thank you :) Swift Qwen 3.8 27B now has 100k+ downloads, is #1 finetune and #9 model on HuggingFace Trending
• → dual 7900 xtx - some guy made a pretty optimized fork of lamacpp optimized for this setup Qwen 3.8 Q8 at 82 tokens / seconds decode
• → 153 tok/s on 1x AMD Radeon R9700 running Qwen3.8 27b NVFP4, 470 tok/s @ 8 conc requests, Prefill @ 3,619 tok/s
• → Qwen3.8-27B at 144 tok/s on an M5 Max MacBook Pro
• → Tuning Qwen 3.8 27B and OMP as a coding agent on 2× 3090s
• → Qwen3.8-Flash-Next (95.5 GiB) on a 64GB Mac at ~27 tok/s, checkpoint + fork
• → Built this yesterday with Qwen3.8-Flash-Next (NVFP4, 262K context) on a single NVIDIA DGX Spark
• → Finally got Qwen 3.8 Next running on my v100 6gpu setup (TP2 PP3)
• → Qwen-3.8-Flash-Next on 1x RTX 5090: TG=50 t/s, PP=2300 t/s - with FreeToken
• → Qwen3.8-Flash-Next at 1M context on Strix Halo: 38 tok/s decode, 18 min prefill (halogen 0.12.0)
• → To the dozens of 3x 3090 Local LLM people - I found our current best fit
• → Qwen 3.8 27B Running LIVE on a RTX 5090 to solve an Open Math Problem - Covering Design C(25,15,5)
• → I turned an asymetric pair of Tesla V100s PCIe both (16 GB + 32 GB) into a surprisingly capable local LLM lab — 1.38k prompt tok/s, 40 decode tok/s with qwen3.8 27B Q6 and Q8...
• → Built a home server from an old PC with GPU upgrade. Qwen3.8 27B runs at ~30 tokens per second.

ROI agen korporat. LangChain membangun agen GTM dan paid media di atas Deep Agents dan melihat konversi lead ke opportunity naik 250 persen, sementara Grab menstandarkan lebih dari 500 layanan agen internal di LLM-Kit sehingga waktu penyambungan layanan baru turun dari dua minggu menjadi satu jam. MCP server WhatsApp milik Meta memungkinkan agen coding mengelola pesan Business, LinkedIn menambahkan konteks organisasi lewat MCP untuk debugging, dan sistem multi-agen DoorDash membersihkan feature flag usang dengan biaya US$4,79 per flag. Madrigal, Abridge, dan Vizient membangun agen layanan kesehatan dengan batasan keselamatan pasien.

• → How We Built LangChain’s Paid Media Agent
• → How we built LangChain’s GTM Agent
• → Grab's Agent Framework LLM-Kit Accelerates AI Agent Production Deployment
• → Meta now lets AI agents handle the boring parts of WhatsApp Business setup
• → Scaling Agents in Healthcare & Life Sciences: Lessons from Madrigal Pharmaceuticals, Abridge, and Vizient
• → Building an Agent Harness for Life Sciences: Introducing Deep Life Sci
• → How Included Health Built Federated Healthcare Agents with LangGraph and Deep Agents
• → DoorDash Uses Multi Agent LLMs to Clean up 60,000 Feature Flags
• → Presentation: Context Engineering at LinkedIn: How We Built an Organizational Context Layer for AI Agents with MCP

Agen coding makin matang. Project HydraFusion di GitHub Copilot, yang masih berstatus preview, mengorkestrasi model secara dinamis untuk tugas coding, sementara Claude Code Projects yang dibangun ulang Anthropic memecah pekerjaan ke sejumlah thread cloud dengan memori dan artefak bersama. Unity merilis plugin resmi untuk Claude Code dan OpenAI Codex, dan MiniMax membuka kode lapisan agen terminalnya dengan lisensi MIT.

• → GitHub Copilot's Project HydraFusion Promises Frontier Level Performance through Multi-Model Routing
• → Claude Code relaunches Projects to manage multiple AI agents in the cloud
• → Anthropic keeps pushing Claude Code toward autonomous coding with new parallel agent workflows
• → MiniMax Code goes open source
• → Unity launches official plugins for Claude Code and OpenAI Codex to stop AI agents from using outdated tutorials

Insiden keamanan agen. Agen AI dipakai untuk spam dan serangan: peneliti memakai Claude untuk menembus GitHub Monorepo milik OpenAI, dan Gemini dari Google menebak kata sandi untuk mengakses tiga perusahaan dalam sebuah uji keamanan. Laporan intelijen AI yang halusinatif nyaris membuat militer AS menaiki sebuah kapal China, dan berkas gugatan hak cipta yang dibuka memuat pernyataan Microsoft yang menyebut pengambilan data AI sebagai "pencurian tenaga kerja terbesar dalam sejarah manusia".

• → The worst spam emails: iLands AI agent hustle
• → AI bots "Timmy," "Ren," and "Jackie" are flooding social media with slop
• → Microsoft exec called AI scraping the “largest theft of labor in human history”
• → Microsoft exec called AI scraping ‘the largest theft of labor in human history,’ new unredacted filings reveal
• → Gemini Hacked Three Companies in First Known Breakout by Google’s AI
• → Security researchers used Anthropic's Claude to hack OpenAI's internal systems in under 72 hours
• → Researchers used Anthropic’s Claude to hack into OpenAI
• → Security researchers used Claude to help them hack into OpenAI
• → Researchers used Claude to hack OpenAI
• → AI hallucination nearly triggers US military operation
• → AI hallucination of Chinese nuclear components almost led to US military attack
• → OpenAI and Microsoft knew they were starting a ‘doom loop’ for the web
• → AI training built on fair use looks shaky when the companies' own people call it "astonishing theft"

Bisnis dan infrastruktur. Anthropic memberi tahu investor bahwa pihaknya akan mencetak kuartal laba kedua berturut-turut dengan pendapatan US$11,5 miliar dan berencana IPO yang bisa menilai perusahaan hingga US$2 triliun atau lebih; Recursive mengumpulkan US$4,65 miliar untuk sistem riset AI. Profound menggalang US$180 juta untuk visibilitas pencarian AI dan Crusoe mengumpulkan US$3,9 miliar untuk pusat data, sementara sebuah jajak pendapat menemukan 61 persen calon pemilih menolak pusat data AI dan BloombergNEF memproyeksikan permintaan gas alam hampir dua kali lipat pada 2035.

• → Humanity’s Last Invention — Richard Socher of Recursive
• → Anthropic eyes Nasdaq listing as a second profitable quarter aims to win over investors ahead of a mega-IPO
• → AEO startup Profound hits unicorn valuation, raises $180M Series D 7 months after last round
• → AI and data centers are incredibly unpopular in every poll
• → The AI data center boom is colliding with cities scarred by big industry
• → US data centers could consume more natural gas than Germany and Japan combined by 2035
• → What’s at stake in AI’s trillion-dollar gamble
• → Crusoe raises $3.9B to build massive data centers and small modular ‘AI factories’