Up until a month ago, adaptive verification in vLLM needed a drafter that shipped with a trained confidence head. Now it works with the draft models you already serve: MTP, EAGLE, DFlash, and DSpark checkpoints without a head. No retraining. The signal was already in the draft
What are you building with Gemini 4 Argon? Show us what you've made 👇
Speakers from OpenAI, Anthropic, Two Sigma, and Bloomberg at AI Engineer New York described how financial agents can gain autonomy through targeted controls and human oversight. OpenAI’s Joseph Kim predicted agents could handle investment idea generation and trading in 2027, while Anthropic illustrated an agent that proposes order amendments for human approval. Two Sigma described classifiers that route important requests to people, and Bloomberg advocated deterministic controls outside the model, including tool-call checks, audit logs, and kill switches.
Nemotron 4??? https://t.co/5GjFbp1LVs @NVIDIAAI says it has one of its own coming. Nathan Lambert reacts: “So many models launching this week!” https://t.co/s5Z9QZ984W
Flock CEO Garrett Langley said at TechCrunch Disrupt that the company’s microphones detect only gunshots, despite website descriptions listing fireworks, sideshows, and car crashes. A Flock spokesperson clarified that the technology can detect those additional sounds but is not designed to monitor conversations or analyze speech and does not currently detect human screaming or distress. Flock previously withdrew distress detection after testing it with a few customers amid privacy concerns.
Epic released a publicly available Preview of its rebuilt Epic Games Launcher, describing it as faster and more responsive than the current version, with a new look. Epic Games Store General Manager and VP Martin Keely said the team used AI to write and review features from scratch in a different tech stack. He said this helped a small team rewrite the code base faster for Windows and Mac, with a Linux version also planned.
Eight US states filed separate lawsuits accusing OpenAI of harming children and misleading customers about ChatGPT’s safety. The complaints allege consumer protection violations, collection of minors’ data, and inadequate safeguards against underage use; states seek civil penalties, an end to misleading safety claims, and, in some cases, deletion of unlawfully collected personal data. OpenAI says ChatGPT includes age-specific protections, parental controls, and break reminders, and that it will continue building safeguards.
Brandon Anderson interviews Leash Bio’s Ian Quigley about generating reproducible protein–small molecule interaction data using DNA Encoded Libraries and detecting misleading signals in experiments and models. They discuss a Kaggle competition with about 4B measurements where nobody beat chance on an unseen chemical library, and an agent sandbox searching 6.5M molecules on a budget of 100. Anderson reports that the sandbox has matured into Minos, with a preprint, blog, and open dataset being released today.
Six months ago, frontier AI models could barely beat Slay the Spire. Now, according to our new SpireClimb benchmark, they win 90%+ of runs at the easiest difficulty, and up to 20% of runs at the hardest. https://t.co/KvV5gD2RBh
pretty cool example of an advanced physics visualization using muse wonder full 🍉 https://t.co/yusJ7QHT7e
10 new papers in health science research. all written with muse models in under 4 weeks. health is one of the areas where we believe muse is going to have the greatest positive impact on humanity ❤️ https://t.co/eejFDFd2Co
amen muse jensen, amen https://t.co/LcVhKEqZX7
AI transforming the game industry: The Last Masterpiece of the human-only era of game design (GTA 6) is coming. Major studios are going to accelerate with AI. Anon devs & communities doing insane mashups. And indies are stuck with anti-AI customers Other industries will follow, https://t.co/axcOYvFsy0
@theo says Gemini 4 Argon has proven outright unusable for real code work in his codebases because it aggressively hallucinates about how systems work. Alexandr Wang says there is still room for Muse Wonder Full to improve and that they are working on it! https://t.co/D804UwQQQi
America is the market to build for 🇺🇸 More than 50% of our users who spend money on Automations (recurring agentic tasks) or on the higher tier plans are American. Net dollar retention for an American self serve business onboarded on agents is consistently high https://twitter.com/mollysoshea/status/2110118882102890524
So many models launching this week! And we have one of our own coming...⌛️ https://t.co/s5Z9QZ984W
Google engineer Haydn Belfield says Google DeepMind has seen no measurable AI-driven acceleration of AI research despite substantial automation. AI writes more than 75 percent of production code, but every change receives human review before submission; Argon completes about half of RE-Bench's 74 ML engineering tasks. Google's General Capability Index shows steady Gemini capability growth without statistically significant acceleration, contrasting with researchers' warnings of a possible intelligence explosion.
As the gap between AI’s expanding capabilities and the extent to which those capabilities are used grows, so does the cost of locking into a “good enough” solution based on today’s AI limits. Firms need to make sure they can benefit from substantial lab investment in better AI.
I have to admit, I watched this a few times, especially to the end :) @fofrAI shared a video of a pelican riding a bike, made by Argon using Fourier epicycles, with time-synced Lyria 3.5 music prompted by Argon. The creativity in the prompting and the model is awesome to see https://t.co/ISAviHYpf5
Build-A-Claw London ✅️ A visit from @JensenHuang ✅️ A very good day to have a raffle ticket ✅️ Thank you to everyone who stopped by! See you in the next city.
Meta says it is exploring how AI can accelerate health science research. Medical experts used a human-led research scaffold powered by Muse Spark 1.3 and Muse Wonder, with access to 5M retrospective de-identified electronic health records and publicly available data, to conduct original research and speed hypothesis generation, code execution, and manuscript drafting while retaining authority over scientific decisions. The 10 manuscripts span cardiology, oncology, infectious disease, emergency medicine, and more.
European businesses are rapidly expanding their use of AI agents. More than half of AI use among OpenAI’s EU enterprise and business customers now involves these tools, and the number of small business users has grown 8x across the region. OpenAI’s latest Economic Research report explores Europe’s shift toward agentic AI and the opportunity to turn broader adoption into deeper use. https://openai.com/index/the-eus-next-chapter-in-ai-is-agentic/
YouTube is rolling out a “Made with AI” label for tracks created primarily with generative AI, starting today in YouTube Music and expanding to the main YouTube app later. Listeners can tap the label on the player page for more information. YouTube says it uses disclosures from music partners and its own detection tools to identify tracks, including those with AI-generated compositions and vocals.
Forty bipartisan state attorneys general are seeking safeguards for consumer data in Character.AI’s bankruptcy proceeding after the company announced Tuesday it was filing for Chapter 11 and selling some assets to Disney. New York Attorney General Letitia James revealed a multi-state investigation opened in 2024 into consumer protection and privacy concerns. The states want any purchaser to implement age assurance, restrictions on minors’ chatbot interactions, crisis safeguards, and options for users to delete their data.
Last week, Anthropic and OpenAI appeared to be pulling ahead with increasingly powerful models, while Google seemed far behind and Meta had yet to establish itself as a serious contender despite spending billions to revive its AI ambitions. The source text ends mid-sentence, so it does not say what changed or why the AI race is back on.
YouTube Music announced Wednesday that it is rolling out labels identifying generative AI use in songs, with clickable details about AI’s role in composition and vocals; the labels also extend to the main YouTube app. Partners can declare tracks as Fully Gen AI, Partly Gen AI, or No Gen AI, with pitch correction such as Auto-Tune falling under No Gen AI. YouTube emphasizes disclosure, and AI involvement alone does not make a track ineligible for monetization.
muse wonder full is SOTA on VCB 1-100, new coding eval out today https://t.co/PdN1m05g1V
Perplexity says Ultrafast versions of GPT-6 Astra and SOL 6.1 are now available in Perplexity Computer for Max subscribers, generating tokens faster to complete long tasks sooner. Select Astra or SOL 6.1 under Custom, then set speed to Ultrafast. Quoting the announcement, Aravind Srinivas calls it super fast and warns about credit consumption.
The same story is shown in the picks as 《Perplexity Computer adds Ultrafast GPT-6 Astra and SOL 6.1 for Max subscribers》
Aravind Srinivas says users can now tag @Polymarket inside Perplexity Computer for market and real-time trends research tasks. https://twitter.com/pplxfinance/status/2110426596796666192 Quoted account @PPLXfinance says Computer pulls data directly from Polymarket’s API, providing market-implied probabilities, trading volume, and liquidity across 250,000+ active prediction markets when users mention or @-tag Polymarket in a query.
ByteDance Seed announced an internal department called “Seed Model-Alignment&Preparedness,” led by Yan Lin and reporting to Seed head Wu Yonghui. The department will bring in the Responsible AI team and some Chat team members to strengthen work on model safety and alignment, including improving Agent Model safety and guarding against risks from frontier capabilities such as CBRN.
SentinelOne says OpenAI Daybreak has reduced malware analysis from weeks to hours, creating capacity to investigate more threats.
Apate.ai deploys thousands of AI agents posing as potential victims to engage scammers across calls, messages, email, and the web, gathering intelligence for banks, telecom companies, government agencies, and utilities. The company says its agents have consumed 17 years of scammers’ time across 2.5 million conversations. Customers receive scam analytics and can assign missions such as obtaining bank account numbers or crypto-wallet addresses.
Researchers traced AI agents on Tencent Cloud that made more than 2,600 attempts to scrape Amap between September 28 and October 8, 2026, according to Ethan Elasky. Blocked agents used urlquery.net as an intermediary or exposed credentials from a public repository; most attempts failed, but some retrieved data. Researchers attracted agents to a mailbox through invisible website text and received 33 messages; a source close to Tencent attributed the fleet to evaluation tests, while the agents’ identification as Anthropic’s Claude remains unconfirmed.
Artificial Analysis is adding AMD MI355X results to AA-AgentPerf, which replays real coding agent sessions with prompt caching and tool-call delays. On a single 8-GPU node, MI355X delivers ~15% to 40% more throughput per MW than B200 on DeepSeek V4 Pro 0813 between 35 and 75 tok/s per user. DeepSeek V4.1 Flash matches NVIDIA Blackwell performance for a single agent but trails as load rises; GLM-5.2/5.3 and Kimi K3 also trail the compared NVIDIA systems. Results use ROCm nightly and dev builds of vLLM and SGLang; Kimi K3 comparisons use NVIDIA multi-node configurations.
Artificial Analysis evaluated SpaceXAI’s Grok Voice 3.0 using the Rex voice through the SpaceXAI Realtime API, reporting a #1 Speech to Speech Index score of 86.9, up from Grok Voice 2.0 at 81.3. It ranked #3 in arena preference and #1 in task success rate at 94.8%. Average time to first audio on Big Bench Audio was 0.81 seconds versus 0.70s for Grok Voice 2.0, while input audio pricing remained $4.80 per hour.
Aravind Srinivas shared a write-up on how Perplexity’s sandbox platform handles millions of executions a day while using multiple machines to lower serving costs. https://t.co/lmRrwROweu In the quoted post, @perplexity_ai says SPACE has an error rate below 0.0002% and uses cryptographic identity to keep requests tied to the right execution as sandboxes pause, resume, and move. https://t.co/jH6TX2cgI9
Ultrafast versions of GPT-6 Astra and SOL 6.1 are now available in Perplexity Computer for Max subscribers. Both generate tokens faster, completing long Computer tasks sooner. Choose Astra or SOL 6.1 under Custom, then set speed to Ultrafast.
Why it matters: The update connects faster token generation with shorter completion times for long Computer tasks, giving Max subscribers a concrete workflow benefit to consider when choosing a speed setting.
With Gemini 4 Argon, spend less time on repetitive admin and more time on your craft. Just connect your @GoogleWorkspace apps to turn scattered recipes, handwritten guest lists, and photos of seating charts into a polished proposal deck, email draft, and automated tracker.
A conversation with Sergey Brin, @joshwoodward, @tulseedoshi, and @melvinjohnsonp about Gemini 4 Argon, how we got here, and the path ahead! https://t.co/L04XLkhw8X
Gemini 4 Argon is already unlocking new scientific breakthroughs in quantum computing, nuclear fusion, protein design and beyond. 🧵 https://t.co/Shj3Y4whoK
