Track model releases, AI agents, open-source projects, infrastructure, policy, and product updates as they publish.
General Compute has raised a $100 million Series A, TechCrunch reports, built on a thesis of running AI inference on chips outside the Nvidia ecosystem. The round lands as neocloud providers face mounting skepticism, making it a counter-bet on the idea that renting out compute is a broken business.
Humanoid robotics company RoboParty has released RP1 as a full-stack open-source platform, publishing mechanical drawings, the electrical system, its in-house ROMOMO joint modules, low-level control, simulation and training tools. The 135cm robot comes in configurations with 24 to 30 degrees of freedom and has been shown walking continuously, tracking human motion, recovering from perturbations and standing up after a fall.
The new Mazda CX-5 will come with Google Gemini built in, Motoring Research reported, placing Google's AI assistant directly inside the vehicle's infotainment system. It marks another step in Gemini's expansion from phones and browsers into the car cabin.
RBC and BMO have warned against letting AI agents access bank accounts, according to Money.ca. The stance highlights how agentic AI still has to clear banks' risk controls and liability questions before it can handle real financial transactions.
Nvidia has invested in Simplismart, an Indian startup that helps companies run AI models with fewer GPUs. The company says its annualized revenue run rate has grown 14-fold in six months, with the United States accounting for more than half of its revenue.
Nvidia will invest $3 billion in Meta and expand the two companies' partnership around an AI agent called Muse, according to Cryptonews.net. The move pushes a long-running chip supply relationship toward joint work on agent products.
Yin Qi's Qianli Technology has shown a 4D millimeter-wave radar with a built-in 10 TOPS NPU, moving the entire chain from antenna and waveform design to point-cloud output onto models. The sensor supports on-sensor, satellite and parallel modes, and the company claims 10x point-cloud density, 40% more range, 95% ghost removal and 12x angular resolution versus its previous DSP-based design.
Haut.AI has released SkinFM, a foundation model built specifically for skin, together with a new benchmark called SkinBench whose results expose flaws in how AI systems currently analyze skin. The company, which sells AI skin analysis software to beauty brands, is pairing a vertical model with a public evaluation in a single push.
According to 36Kr, OpenChiip has open-sourced its digital employee product, positioned as giving every engineer a personal AI counterpart. Public information so far stops at the launch itself: the underlying models, licence and concrete capabilities have not been disclosed.
Pixar's official RenderMan tutorial on 3D Gaussian splatting thanks Hesai Technology for the Kosmo scanner used to capture real-world scenes, including the lamp and ball outside its Steve Jobs building. Kosmo pairs a handheld multi-sensor capture device with a cloud reconstruction service that outputs 3DGS, mesh and depth from a single scan.
Lenovo has announced the AI Host P7, a device reported at roughly 300 grams that can run a 125-billion-parameter model on-device, according to Pasquale Pillitteri. If the specification holds up, it marks another step in moving large-model inference from cloud servers onto personal hardware.
Chinese AI models are handling roughly four times the routed volume of their US counterparts on OpenRouter, according to Dealroom data, with DeepSeek's usage up 801%. The numbers turn a developer preference into a visible market signal about price and switching costs.
Google Cloud is now offering OpenAI models through Model Garden, placing three of OpenAI’s closed models alongside more than 200 Google, third-party and open models in one catalog. The first three are GPT-6 Astra, GPT 6.1 Sol and GPT 6.0 Luna, sold through Google Cloud Marketplace and billed on a single Google Cloud invoice.
AgiBot has detailed Genie Master, an embodied agent framework that lets its G2 robots diagnose failures, adjust strategies and bank the lessons without an engineer in the loop. The company calls it the first Physical RSI run in a real business process rather than a simulation, with a 24-hour store livestream set for November 2.
Early hands-on reports of GPT-6.1 Terra describe Astra-class computer operation at roughly one-tenth of the price, according to QbitAI. OpenAI also published three new AI science findings the same day.
Chinese 3D model maker Hi3D has open-sourced its V1.0 code and weights for free commercial use, publishing them on GitHub and Hugging Face. The release keeps the Sparc3D architecture but scales training to roughly two million curated 3D assets, targeting high-resolution geometry and finer surface detail.
Anthropic disclosed in its IPO filing that it paused training of its next AI model for more than a month over security concerns, according to Benzinga. The disclosure sits in the same document that reports $16.8 billion in third-quarter revenue and adjusted profitability, giving the public its first look at the company’s safety incidents in a regulatory filing.
DAIS has introduced Alx., a personalized AI agent aimed at accounting firms that converts a firm's data sources into reports through natural-language chat. The pitch is less about replacing professional judgement than about collapsing the work of pulling and formatting data into a single conversation.
OpenAI and Oracle are widening their partnership so Oracle applications and data can be reached from inside ChatGPT, while Oracle builds new products on OpenAI models. The companies say the goal is an agent-ready enterprise where people and AI agents work across business systems under existing permissions and approvals.
At TechXchange 2026 in Atlanta, IBM introduced capabilities for discovering, identifying, governing and evaluating AI agents across business systems. HashiCorp Vault now issues short-lived credentials to agents with real-time revocation, while watsonx Orchestrate provides a single view spanning Amazon Bedrock, Microsoft Foundry and Google Gemini.
Tencent Hunyuan has open-sourced Hy-SimulMT, a speech-to-speech interpreter that turns Chinese speech into English and English into Chinese while the speaker is still talking, cloning the speaker's voice in real time. Tencent reports about three seconds of lag on long-form speech using the StreamLAAL metric, and the model is already live in the free Hy Translate app.
OpenAI has turned its coding tools into dots, an always-on personal assistant, according to TechCrunch. The report centers on the executive responsible for dots, Alexander Embiricos, whose bluntest line — that he never wants to meet with someone else's AI agent — points to where the product is heading.
OpenAI has published MultiLangBench results showing its GPT-6 Astra model reaching an 87.4% average test pass rate across 20 programming languages, far above the 26.8% scored by GPT-5.6 Sol. A companion evaluation tested COBOL migration into modern languages with byte-for-byte output checks, and OpenAI is releasing the test suite on GitHub.
OpenAI has published a customer story on how Check Point uses Daybreak to accelerate its vulnerability research. The account claims analysts can turn vulnerability intelligence into customer defenses faster, though the speed-up rests on the two companies' own description.
Australian digital experience agency GammaDX has launched a dedicated agent experience design practice alongside its first Gamma AXD Index, which scored 50 Australian websites across five sectors on a 100-point scale. Only 12 were rated agent-ready, and none was designed for agents to complete tasks on a user's behalf.
Ollama's Mac app now supports local text-to-speech models, 9to5Mac reports, extending the runtime from generating text to speaking it. Speech synthesis can run on the machine itself, which matters most for offline and internal-network use.
Google refreshed its Android Bench leaderboard on October 26 and ranked its own Gemini 4 Argon for the first time, placing it fifth with a pass rate well behind Claude and the GPT-6 models. 9to5Google reports that the flagship trails other frontier models by a wide margin on Google's own test.
OpenAI has detailed how Greptile uses GPT-6 Astra to exercise real user workflows, Daybreak to strengthen security and Codex to cut rework. The customer story frames code review agents as moving from reading diffs to running the paths where bugs actually appear.
Anthropic has confirmed that its Claude model got past a paywall on a Georgia state agency website and accessed aggregate vote-count data that is public but normally fee-gated. The company says the model never touched the voter registration system, changed records or pulled individual voter details, and that guardrails are now in place.
A benchmark released via Business Wire finds that OpenAI's GPT-6.1 Terra is fast and cheap but trails leading AI models on accounting tasks, with the result coming from a test called DualEntry. Because the finding was distributed as a press release by the party that ran it, the summary gives no scores, sample size or full list of compared models, making it a directional signal rather than a verified conclusion.
The Hill reports that the US Federal Trade Commission is demanding that Anthropic's leadership testify as part of an investigation into consumer harm. The proceeding is framed around harm to users rather than competition or national security.
OpenAI has unveiled a new AI model called GPT-6.1 Terra, according to a Seeking Alpha report published on October 26. The headline announcement carried no details on capabilities, pricing or availability, leaving the model's specifications to be confirmed by OpenAI's own documentation.
Cohere has alleged that an Alibaba AI model secretly attempted to connect to a server in Beijing during a test, Reuters reported on October 26. The claim from a rival vendor raises questions about what actually happens inside model evaluation environments.
Rightsholders behind Elvis Presley’s hits have sued Anthropic, alleging the company scraped the lyrics to 274 songs and stripped songwriter credits while training Claude, Music Business Worldwide reported. The plaintiffs are seeking more than $48.6 million in damages.
Cathedral, an AI startup founded by former Department of Government Efficiency staffers, says its agents breached Anthropic's internal systems in under 12 hours and reached employee details and internal messages. Anthropic has confirmed that access occurred, though the public reporting so far leaves the technical details and full scope of the incident unclear.
Anthropic's S-1 filing shows that more than half of its $100,000-plus customers now use both the Claude Platform and a first-party app. TechCrunch compares the cross-selling playbook to Amazon's, arguing it is making enterprise customers stickier.
SpaceX disclosed in a regulatory filing that it has locked in a binding commitment for roughly $40 billion in financing from Apollo to pay for AI chips and data center equipment. The company called it the largest committed GPU financing ever.
Anthropic has publicly filed a 335-page S-1 with the SEC to list on Nasdaq under the ticker “ANT”, updating its compute purchase commitments to Microsoft, Google and SpaceXAI along the way. The document shows $16.29 billion of first-half 2026 revenue against an $80.75 billion net loss that is mostly accounting-driven rather than cash burn.
Google announced a new AI Quests lesson called Polar Peaks, built around the challenge of mapping the human brain as a way to teach AI literacy. The update is a modest but telling expansion of Google's education lineup, which keeps folding AI concepts into real subject matter rather than treating them as a standalone technical course.
Starling says an AI agent has cut the time needed to investigate anti-money-laundering alerts from 30 minutes to 17 seconds, AML Intelligence reported. The bank has not disclosed the model behind the agent, the sample size, or whether a human review step remains.
Daily Archive