All AI news

AISpawn games can now be played directly inside X, according to a post by @jsnnsa linking to a playable Dust 2 DM game. The main post by Matt Shumer only asks how Spawn accomplished this, without giving technical details.

AIArtificial Analysis reports that Grok Imagine Video 1.5 Lite comes closest to the frontier on AA-Video-T2V v2.0 in Multi-Scene & Narrative, Lighting & Materials, and Text Rendering. It is furthest behind in Dialogue & Lip Sync and Human Anatomy. Compared with Grok Imagine Video 1.5, Lite matches it in Physics and trails on the other nine capabilities, by the least in Multi-Scene & Narrative.

AIAmong 12 models on AA-Video-T2V-Silent v2.0, Grok Imagine Video 1.5 Lite is the only one that is both fastest and highest quality, with no model beating it on both measures. It generates a 10-second 1080p clip in a median of 60.5 seconds. Kling 3.0 1080p (Pro) scores slightly higher but takes 94 seconds for a 5-second clip, while Vidu Q3 Turbo is 9 seconds faster on a 5-second 720p clip yet scores well below it.

AIGrok Imagine Video 1.5 Lite ranks #17 on AA-Video-T2V v2.0, two places above Google's Veo 3.1. At 1080p with audio, it costs $0.14 per second versus $0.40 per second for Veo 3.1. Compared with Grok Imagine Video 1.5, Lite is 44% cheaper at 1080p but ranks six places lower.

AISpaceXAI's Grok Imagine Video 1.5 Lite ranks #17 on both AA-Video-T2V v2.0 leaderboards, ahead of Google's Veo 3.1 at about a third of its price. It is the fastest model at its quality level in Artificial Analysis benchmarks, with a median of 60.5 seconds for a 10-second 1080p clip, and it costs $0.14 per second at 1080p, 56% of Grok Imagine Video 1.5's $0.25 per second.

AICNBC's Jim Cramer said Thursday's AI stock sell-off, which followed a Financial Times report that OpenAI's annualized revenue was about $20 billion below previously signaled levels, shows why investors need diversification. Oracle fell 5.5% and Broadcom dropped 4.35%, while Home Depot rallied as Treasury yields eased. Cramer argued that concentrated portfolios risk panic-selling and missing a recovery.

AIAnthropic published a guide on building automations with Claude Managed Agents, which can be deployed to read Slack and GitHub on a schedule, with credentials and memory, and post updates. The setup can be started with a Claude Code command, and Max and Team plans include new API credits to use it.

AIElad Gil posted an enthusiastic reaction to Hone's blog post "Closing the Gap," which discusses improving AI's real-world value. The post is brief and offers no concrete product details, figures, or benchmarks beyond the linked material.

AIBaiju Bhatt, founder and CEO of Cowboy Space, says the company is developing an energy grid in orbit. Its first satellite, Reason 1, reached orbit on a SpaceX rideshare, and the team is communicating with it daily.

AIThe AI-mperor has no clothes

AIReplit says users can keep working in the same conversation to request a pitch deck or launch graphics after building a website. The post notes the brand and direction are already established, so the work can build on them without starting over.

AIReplit says users can extend a survey analysis in the same conversation by requesting an action plan, a presentation, or a research brief. The next step builds on the findings already explored in the chat.

AIDatabricks has released Funke, a Python and PySpark library and deployable pipeline that parses HL7v2 healthcare messages into native Spark types while preserving the full message hierarchy. It succeeds Smolder, the Scala data source Databricks open-sourced in 2021, and ingests through Auto Loader into Unity Catalog bronze and silver tables. Users can query segments, fields, components, and subcomponents directly with DataFrame or Spark SQL expressions.

AIDatabricks introduces database branching in Lakebase Postgres, letting each coding agent work in its own isolated database branch created in under a second regardless of size. Branches use copy-on-write storage, consuming extra space only as they diverge, and scale to zero when idle so unused branches incur no compute cost. Schema changes are tracked in code and promoted to the parent branch through migrations rather than merged back, and ephemeral branches are created per pull request for testing.

AIAlphabet spinoff Isomorphic Labs is reportedly in early-stage talks to raise a new funding round at a valuation of $40 billion to $50 billion, according to Bloomberg, which did not specify the round's size. Isomorphic's flagship AI platform, IsoDDE, is designed to automate binding pocket discovery, which it says is more than 50% better than the closest competing tool, and it models antibody interactions about 2.3 times better than AlphaFold. The company earns revenue through research partnerships with Eli Lilly, Novartis and Johnson & Johnson.

AIAnthropic has launched OSS Scanner, a free opt-in service that gives open-source projects periodic security scans by its strongest models, including Claude Mythos. The reports are fully model-generated without human review or triage, so they may be incorrect or invalid.

AIThariq, who works at Anthropic, accidentally made public the GitHub repo for ai-newtab, a Chrome extension he uses daily. It replaces the new tab page with an AI-generated homepage built each day from sites he reads.

AIFrançois Chollet argues that most AI progress measures from 2023 to 2026 fit an exponential curve, while AI capital expenditure over the same period has grown slightly super-exponentially. He concludes that such externally funded growth is unsustainable, so sustained exponential progress will require matching exponential profits to close the resource loop.

AIArena's account of its continued work with The House Fund is a short acknowledgment with no product, figures, or new details. Read alongside the linked background, it sits in the context of Arena's $200M Series B at a $3.1B valuation and its Alignment Index for frontier models.

AIGergely Orosz argues that people chasing AI tools and scheduling software to look credible or influential are seeking shortcuts that don't exist. He says trust in a person or company cannot be gained through shortcuts, building on Lulu Meservey's point that credibility has become the scarcest resource.

AIPaige Bailey, who works on Google and Gemini, argues that building tools for humanity requires building them in collaboration with people. The quoted post from Pushmeet Kohli says Google DeepMind has focused on deep collaboration with the mathematical community to advance mathematics.

AIArena, which began in a small room in Berkeley's Soda Hall and now has about 90 people, is hiring across ML, product/engineering, design, sales, and more as it scales. Candidates can DM @aryanvichare10 or apply through the link in Arena's bio.

AISalesforce Ventures announced an investment in Arena's Series B round, with Arena described as a neutral platform where real users vote on anonymous model outputs across text, coding, vision, image generation, and agentic tasks. Labs and builders use its leaderboards to guide training and purchasing decisions.

AIOpenAI's YouTube livestream presents three demos showing how Codex supports production monitoring. The demos cover tracing a checkout failure in Grafana, investigating a Kubernetes rollout causing out-of-memory restarts, and linking a Codex Security finding about a missing resource limit to service availability.

AIPrima Mente and AlzData are launching the Alzheimer's Translation Challenge, a global AI competition to discover new therapeutic hypotheses for Alzheimer's disease. The challenge centers on a 150M cell atlas of neurons, astrocytes, and microglia across different genetic backgrounds under combinatorial perturbations, with multi-modal readouts. Top teams will have their hypotheses tested in Prima Mente's wet lab, and the data will be available through the AD workbench, Hugging Face, and Prima Mente's modeling platform.

AIArena announced a $200 million Series B at a $3.1 billion valuation and launched the Arena AI Alignment Index, a new measure of how AI agents behave in the real world. The main post is a thank-you from Felicis, a backer of Arena since its seed round, and the company's origins as a UC Berkeley research project.

AIFireworks announced that Dylan Patel, founder and CEO of SemiAnalysis, will speak at its Forge event on optimizing AI infrastructure for inference, from chip level up. The event takes place November 3 in San Francisco.

AICoreWeave launched Forge, a platform meant to help teams run, observe, evaluate and improve AI agents continuously through a five-stage loop. The offering includes Agent Lens, Registry, RL Rollouts, model distillation, and the AI Research and Iteration Agent (ARIA), which is now generally available. CoreWeave also introduced a partner network of tested, co-engineered integrations, according to Susanne Seitinger, vice president of product marketing.

AIApplied Digital shares fell about 2.5% after first-quarter revenue rose more than 320% year over year and beat estimates. CEO Wes Cummins said the gap between revenue growth and the share reaction should narrow over the next 12 months. He expects all contracted capacity to be online in roughly 20 to 22 months.

AIArena has released its Alignment Index, a ranking that tracks AI models for unauthorized actions, deceptive completions, and false attribution of intent. The full technical report and rankings are available on Arena's blog and leaderboard site.

AIThe startup behind a popular AI model leaderboard has raised a new funding round at a $3.1 billion valuation. The main post points readers to Bloomberg's coverage of the round for further details.

AIMidjourney is testing a new "thinking mode" for its image generation on its Alpha website, alpha.midjourney.com. The company says the mode improves prompt accuracy, typography, and coherence, and it is asking users to try it and share feedback.

AITogether AI, n8n, and NVIDIA speakers will present at Open Haus Berlin on building with open frontier models, with talks by Riqwan Thamir, Harry Kim, and Zain. The event is held in a digital art museum and includes food, drinks, and discussion of evals.

AIMIT Statistics and Data Science Center director Sasha Rakhlin argues that AI progress is fastest where results can be verified quickly, citing a model reaching gold-medal level at the International Mathematical Olympiad a year before models produced new research results. He says departments should reconsider how they reward work, emphasizing question-asking, replication, and disclosure of AI's role in a researcher's contributions. He also urges universities to build shared lab infrastructure that captures failed experiments and tacit expertise.

AIOpenAI told investors it reached roughly $50 billion in annualized revenue at the end of September, below the $68 billion figure widely reported last month. Nvidia, Oracle, and CoreWeave shares fell on Thursday, with a person familiar saying the $68 billion figure included partner gross revenue. OpenAI is also weighing a possible funding round of around $30 billion and has not finalized a term sheet.

AIEpoch AI reports that GPT-6.1 Sol's latency behavior on long contexts suggests a change in how the model handles them, though not conclusively a new architecture. The post is a follow-up to its earlier report on latency scaling in frontier models.

AIEpoch AI reports that OpenAI halved GPT-6.1 Sol's cached-input price compared with GPT-6 Sol. Its measurements also show the model handles long prompts faster. The post raises, but does not confirm, a new architecture for GPT-6.1 Sol.

AIMidjourney is testing a "Thinking Mode" on its Alpha website, where users can click "Rerun (Thinking)" in a job's lightbox to regenerate an image. The company says early tests show gains in prompt accuracy, typography, and coherence, and it is asking users to share feedback in its #ideas-and-features channel. It may later offer the mode broadly or as an option to add more thinking after a job.

AIMidjourney's alpha site now lets users share folders with others as Collaborators or Viewers, with sharing by link also available. A new thinking mode lets users rerun jobs made with 8.2 standard and edit models to fix missed prompt details such as objects, layout, anatomy, and text. Sharing does not change image privacy, so non-stealth images can still appear on Explore and profiles.