[AINews] TypeSafe/Jev at >$100M ARR, $7.5B valuation 3 weeks after launch
Latent Space AINews reports that TypeSafe's Jev surpassed $100M ARR three weeks after launch, reaching a $7.5B valuation, as decision models emerge as a product category.
Today 0 items
Latent SpaceAI score5252 Latent SpaceAI score4747 Why AlphaFold Didn't Solve Protein Folding — Pushmeet Kohli, Google DeepMind & Sal Candido, Biohub
Google DeepMind's Pushmeet Kohli and Biohub's Sal Candido discuss why AlphaFold's breakthrough was only a starting point and protein structure prediction is far from solved.
MIT Technology Review · AIAI score6565 We’re putting too much faith in AI’s ability to say no
MIT Technology Review reporter Arthur Holland Michel writes that refusal has become the load-bearing wall of AI safety, but the mechanism is inherently probabilistic and not fully understandable, and jailbreaks keep emerging: less than three days after Anthropic Fable 5's release, Amazon researchers unlocked some of its hacking capabilities.
Latent SpaceAI score4343 [AINews] not much happened today
OpenAI fired three safety researchers — Tomek Korbak, Mikita Balesni and Jasmine Wang — who say they were let go for 'putting safety above the company's short-term interests.'
TechCrunch · AIAI score6666 Google brings agentic AI to Gemini, starting with businesses
Google announced at a Google Cloud event a unified agent for Gemini that can accept goals rather than instructions, autonomously plan tasks, use skills and tools, and connect to enterprise internal systems.
The Verge · AIAI score5555 Google’s AI note-taking app transcribes your meetings completely offline
Google has released Google AI Edge Foresight, an experimental note-taking app that transcribes meetings and audio files completely offline for free, running on macOS via the on-device EmbeddingGemma 2 model. The app can jot brief points during meetings and generate full notes, and a built-in assistant can answer questions about the content with citations to local files; Google says files, audio, and notes never leave the computer, and it is currently optimized only for Apple Silicon Macs.
The Verge · AIAI score5151 Google is launching a one-stop Gemini agent for your work tasks
At its Gemini at Work event, Google launched a general-purpose Gemini AI agent that lets users chat and assign tasks through a single interface inside the Gemini Enterprise app.
The DecoderAI score7575 One public-facing AI agent on AWS could read, rewrite, and delete every other agent in the region
Zenity Labs researchers report a vulnerability chain in AWS Bedrock AgentCore, dubbed AgentCorruption, where a single prompt sent to one public agent within the same AWS account and region can take over all AgentCore agents in that region.
TechCrunch · AIAI score5454 Google releases a new local-first Granola competitor
Google AI Edge Foresight, a Mac app from the same team, is a Granola-style meeting notes app that runs entirely offline and is optimized for Apple Silicon.
MIT Technology Review · AIAI score6565 AI breakthroughs in robotics won’t change your life any time soon
In a piece produced with Aventine, MIT Technology Review reports that despite optimism about humanoid robots from Musk, Jensen Huang, and others, robotics researchers broadly believe current AI methods can't support general-purpose robots.
Google ResearchSelectedAI score7272 Does better work always mean better workers?
Google Research published an NBER paper reporting a three-month randomized field experiment with 133 lawyers at 11 IP firms, giving two-thirds of the lawyers early access to the then-unreleased Google Labs AI patent drafting tool (now part of Gemini Notebook).
Why it matters: The three-month randomized experiment separates AI-assisted from unassisted performance, and readers can see the transferable finding that junior lawyers' skills did not grow along with output.
The DecoderAI score5858 Google says 180 billion images and videos now carry SynthID watermarks as detector goes public
Google has opened SynthID Detector to the public, a tool that detects invisible watermarks in content generated by models such as Gemini, Veo, and Lyria, supporting JPG, PNG, MP4, and MP3 formats.
The DecoderAI score5858 Zuckerberg's Biohub leads a $1.8 billion push to build AI models that predict cell behavior
Biohub, the nonprofit funded by Zuckerberg and Priscilla Chan, is leading a $1.8 billion effort to train AI models that predict cell behavior to speed up drug development, including $500 million from the five-year Virtual Biology Initiative they pledged this April.
The DecoderAI score6161 Google bets Gemini can turn casual players into game developers with new Playground feature
Google launched Playground, a browser-based AI platform where adult US users can create games from text alone, with no coding skills required. Powered by the Gemini, Nano Banana, and Lyria models, it lets users conversationally adjust game rules, physics, characters, and environments with instant testing, share creations via link or publish them to a public gallery, and some genres support leaderboards and multiplayer.
TechCrunch · AIAI score5555 Google experiments with an AI-powered gaming platform
Google Labs launched AI game creation platform Playground on Wednesday, letting users create browser-based games from text prompts with no coding skills. Users can pick genres like puzzle-quiz or racing, 2D or 3D, single- or multiplayer, describe gameplay mechanics and visual style, and upload images that AI turns into game assets; finished games support leaderboards and can be saved privately, shared via link, or published to the Explore gallery.
The Verge · AIAI score5454 Google invests millions in Mark Zuckerberg’s efforts to create a ‘virtual cell’
Google DeepMind, Meta, and AI drug discovery startup Isomorphic Labs are jointly investing $300 million in Biohub, the nonprofit biomedical research organization founded by the Zuckerbergs, according to Reuters.
Ars Technica · AIAI score6262 Google rolls out improved SynthID AI content detector, now available globally
Google announced the SynthID detector now supports watermarks from all partners and is open to everyone at the new site SynthID.com. Previously external users could only check watermarks via Gemini, and the old detector only recognized Google's own SynthID, so content like ChatGPT images carrying SynthID wasn't flagged.
TechCrunch · AIAI score6565 Google’s new SynthID website can identify AI-generated media
Google launched a SynthID verification website where anyone can upload images, video, or audio to check whether it was AI-generated; the tool was previously in testing with select journalists, media professionals, and researchers. The site supports image formats like JPG, PNG, WEBP, and GIF, video formats like MP4, MOV, and WEBM, and audio formats like WAV, MP3, OGG, FLAC, AAC, and M4A.
Latent SpaceAI score7272 [AINews] Quasi-Riemann-Hypothesis: OpenAI publishes 722 math papers solving 90 of the top 500 open math problems; “the most significant moment” in >100 years of mathematics
Latent Space released its AI news roundup covering October 5-6, 2026, headlined by OpenAI publishing 722 math manuscripts produced by an internal frontier model, alongside Mistral Large 4 and several open-source model launches that week.
Ars Technica · AIAI score6565 AI glasses face their first major government crackdown
Australia is considering banning smart glasses in certain public places, and courts in several countries have issued bans; Norway's government calls a ban a high priority, with a draft bill to be submitted to parliament soon, though it is not seeking a blanket ban. Norway's Micaelsen said people should not have to worry about hidden recording, photos and filming in settings unaccustomed to surveillance.
Ars Technica · AIAI score6363 MCP for agent-to-agent comms may be the riskiest protocol you've never heard of
According to Ars Technica, over the past five months Google and four other organizations have acknowledged vulnerabilities exploiting MCP: an attacker hijacks one agent inside a target network and uses chained prompt injection to spread malicious instructions to other internal agents, such as stealing databases and sensitive information.
The DecoderAI score7272 Google's new image model Nano Banana 2.1 generates better images for less money
Google has released Nano Banana 2.1, an image generation and editing model that succeeds Nano Banana 2 (released in February 2026), at roughly half the price: 1K images drop from 6.70 cents to 3.36 cents, and 4K from 15.10 to 7.56 cents.
The DecoderAI score6161 Google claims EmbeddingGemma 2 outperforms rival embedding models twice its size
Google has released the open embedding model EmbeddingGemma 2, which converts text, images, video, audio and code into vectors. With 740M parameters, Google says it beats rivals twice its size on multimodal embedding benchmarks.
Google ResearchAI score4747 Open and Emergent Problems in Agentic Privacy and Security: A Contextual Angle
Google released a report from its CAPS workshop, bringing together more than 50 academic and industry leaders to discuss the privacy and security challenges of autonomous agents. The report notes that agents differ from traditional software in three ways—unstructured interfaces, probabilistic control flow, and autonomous delegation—and, building on Contextual Integrity theory, proposes contextual policy engines combined with system-level sandboxing, model-level reasoning, and user-centric controls as a multi-layered defense.
Google ResearchAI score5252 Unlocking Earth AI’s planetary geospatial foundation models for global public health
Google Research opened up location embeddings from Earth AI's Population Dynamics Foundation Model (PDFM) as plug-and-play inputs, completing independent validation with five institutions across five types of public health challenges, covering MMR vaccination, cardiovascular disease, dengue, postpartum depression, and cholera.
Simon WillisonAI score2323 Mistral Large 4
Simon Willison tested Mistral Large 4 with an SVG generation prompt about an armadillo in fishnet stockings jaywalking on Mars, comparing its output with claude-opus-5.5, gpt-6.1-sol, and gemini-3.8-flash, with each model using its default reasoning level.
The Verge · AIAI score2424 Gemini Call for Me might tell your mom you’re running late
An Android Authority APK teardown uncovered a "Gemini Calling" onboarding screen with examples like "call my mom and tell her I'll be 15 minutes late," suggesting Google may extend "Call for Me" from business calls to leaving messages for friends and family.
The Verge · AIAI score6666 Google is about to remove free access to Gemini Flash and Pro
Google announced that from October 9, free Gemini users will only get the Flash Lite model, while the standard Flash model will require the $4.99/month Google AI Plus subscription.
The Verge · AIAI score5151 We can’t just change the definition of ‘recording’
The Verge writer Victoria Song criticizes tech companies for pushing to redefine recording on AI hardware.
Google DeepMindSelectedAI score6868 EmbeddingGemma 2: an open, lightweight multimodal embedding model
Google DeepMind released EmbeddingGemma 2, a 740M-parameter model built on the Gemma 4 architecture under the Apache 2.0 license, mapping text, code, images, video and audio into a shared embedding space.
Why it matters: The official release provides concrete metrics such as parameter count, memory footprint and benchmark scores, helping readers judge whether it suits on-device retrieval scenarios.
Latent SpaceAI score3434 [AINews] not much happened today
OpenAI released GPT-6.1 Sol, priced at $2/$10 per million input/output tokens, ranking 5th on the Agent Arena, 39% cheaper than GPT-6 Sol with a score 1.52 points higher, and 81% cheaper than Astra.
Google ResearchAI score5151 Toward provably private learning from federated data
Google has announced a next-generation federated learning system based on trusted execution environments (TEEs), providing remotely verifiable and auditable guarantees for data anonymization. Access policies are published to the public transparency log Rekor, the KMS and data processing binaries are reproducibly built from open source code, and encrypted training data is decrypted and processed only inside the TEE, with workload operators seeing only metrics and differentially private model weights.
TrendForce Insights Semiconductor Industry InsightsAI score6262 Google Goes Global. AWS Has the Most GW under Construction
TrendForce released a monthly report based on its Data Center Tracker (covering nearly 2,000 facilities), with AWS adding the most MW this month, contributed largely by the Susquehanna campus accounting for nearly half; Google adjusted across three countries, led by 330 MW of new air permits in Omaha.
MIT Technology Review · AIAI score5656 Don’t be fooled—LLMs don’t reason
Thore Graepel, a former core member of DeepMind's AlphaGo team, argues in an essay that large language models are essentially fast-associative next-token prediction, that their chains of thought are not an independent reasoning mechanism, and that they do not reason in the scientific sense.
Ars Technica · AIAI score7171 Judge dismisses Chegg and Penske antitrust lawsuits targeting Google AI search
US federal judge Amit Mehta dismissed the antitrust lawsuits by Chegg and Penske Media against Google, ruling that Google's conduct in AI overviews and similar products does not violate antitrust law.
The Verge · AIAI score4747 Google’s new Guided Vision feature can help you read the fine print
Google today launched Guided Vision in Gemini Live on compatible Android devices, providing real-time voice descriptions of what the phone camera sees, helping users read fine print, identify objects, and describe their surroundings.
TechCrunch · AIAI score6363 Google thinks SpaceX’s Starship has to launch 1,800 times before space data centers get off the ground
Google's Project Suncatcher prototype satellite launched on October 1 aboard a SpaceX rocket from California, putting its TPU in space for the first time; built by Planet Labs, the satellite will run TPUs in 15-minute bursts to verify in-orbit operation.
TechCrunch · AIAI score6868 Opus 5.5 loves to tell you ‘this matters’ (and other AI writing tells)
TechCrunch reported on Graphite's new research: comparing 10,000 human articles written before ChatGPT's release with rewritten versions by various models, it identified 13,000 phrases that appear at least 2x more frequently in AI text.
The Verge · AIAI score6767 Judge dismisses antitrust lawsuits over Google’s AI Overviews
US federal judge Amit Mehta on Wednesday dismissed two antitrust lawsuits against Google brought by Chegg and Rolling Stone parent Penske Media, which had accused Google of forcing publishers to provide content to AI Overviews for free or vanish from search results.
TechCrunch · AIAI score5353 America.gov gets really weird when you ask it about Minecraft, but it’s not a glitch
The US government launched America.gov, a public chatbot built with help from Google and SpaceXAI, which is currently hard to jailbreak and confirms that Biden won the 2020 election.