All AI updates

How Agent Anomaly Detection scores on benchmarks we didn't write

Google tested its Private Preview Agent Anomaly Detection feature on five public, non-Google agent-security benchmarks (AgentDojo, InjecAgent, Agent Security Bench, AgentDyn and AgentHarm), running 1,774 sessions through an ADK agent using gemini-2.5-flash. It flagged 87% of harmful instructions the agent actually carried out, with false alarms on 2% to 16% of benign sessions, but caught only 4 of 18 sessions where poisoned long-term memory drove the agent to act. The post details per-benchmark and per-app results, known gaps such as resisted instructions and small actions, and recommended defenses including Model Armor, Semantic Governance Policy and human approval for high-impact actions.

The AI 100: The 100 Private Companies Defining the Age of AI

Newcomer, in partnership with Bessemer Venture Partners, released its inaugural AI 100 ranking of the most promising private AI companies, with a collective valuation of roughly $3.5 trillion, led by Anthropic at #1, OpenAI at #2, and Databricks at #3. Researcher-heavy model companies make up only 9% of the list but 55% of its valuation, while inference players, pivoted startups like Stripe and Anduril, and video, coding, legal and robotics firms also feature prominently. The ranking weighed revenue, valuation and market share alongside people and culture, with figures fact-checked against company data and PitchBook.

0.170.0-alpha.4

This entry is a GitHub Changelog page for version 0.170.0-alpha.4, but the supplied material contains only site navigation and product menu text, with no release notes or details about the version itself.

Eclipse Aerospace simplifies flight-data ingestion with Databricks

Eclipse Aerospace worked with Databricks to replace a multi-stage AWS Lambda and Avro pipeline with an Arrow-native PySpark custom data source that wraps its existing Rust flight-data decoder, reading proprietary telemetry files directly in Spark on serverless compute. The approach, enabled by the Python Data Source API available from Databricks Runtime 15.4 LTS, cut latency, removed duplicate Avro storage and a Lambda tier, and brought decoding into the Spark UI and job metrics. Eclipse reports the connector can process records within a minute of upload and commonly handle multi-gigabyte batches in parallel in under two minutes.

How Constellation is building a governed data foundation on Databricks

Constellation, the largest US nuclear fleet operator, is building a governed Enterprise Data Platform on Databricks to give data consistent shared meaning across domains, choosing Databricks partly because it interoperates with existing architecture rather than requiring a rip-and-replace. The platform rests on conformed semantic definitions, a semantic knowledge layer with selective Data Vault modeling, and Unity Catalog for lineage, fine-grained access and ownership of meaning, with early use cases including a nuclear engineering assistant built with Databricks Genie, outage planning, and sales intelligence.

Show HN: Headwig - Run and test parallel AI agents without Git worktrees

Headwig introduces "Wisdom" groups that bundle an app's services so they can be built and run together, such as a frontend and backend as two services in one Wisdom. Users create a Wisdom via New Wisdom, then add services by clicking the three dots beside the Wisdom and choosing Add Service. The post also points readers to Discord for questions.

OpenAI, Google, Anthropic hit with lawsuit to stop AI development

Reuters reports that OpenAI, Google and Anthropic have been hit with a lawsuit seeking to halt frontier AI development. The article is dated 2026-11-12.

AI search is recommending fake retail websites

The study collected 6,795 answers from the logged-out, consumer-facing versions of ChatGPT, Gemini and Perplexity, working with Gauge, an AI search company. The material is largely a FAQ for Podqi, an AI-native IP enforcement service that detects misuse across 180+ platforms and files takedown notices, typically resolving within 48 hours.

I Made an AI Animated Short Film: Here Is the Full Process [video]

A video walkthrough of the full process behind an AI-generated animated short film, shared on YouTube and submitted to Hacker News. The submission had 1 point and no comments at the time of collection.

Nectar launches an AI agent to automate creator discovery and outreach for brands

Nectar, an AI-powered marketing operating system founded by former Meta employees Misbah and Farah Uraizee, has launched a "Creator Agent" that searches over 300 million creator profiles, checks content and prior brand partnerships, sends personalized messages in a brand's voice, and helps manage contracts, gifting, and shipping. The agent, available globally for Instagram and TikTok creators, also drafts outreach, builds custom landing pages, and tracks campaign performance, with planned availability through MCP for assistants like Claude, ChatGPT, Gemini, and Copilot. Nectar raised a $30 million Series A led by Menlo Ventures and its Anthology Fund in May.

Google’s latest Search upgrade promises better answers to your most complicated questions

Google says its new Gemini 4 Hydrogen model, described as its "biggest model improvement to Search yet," improves Search's handling of multi-part questions, highly specific requests and confusing typos. Hydrogen completes tasks 2.5 times faster than Gemini 3.8 Flash and is rolling out to all Search users globally, while also coming to AI Studio, the Gemini API, the Gemini App, AI Pro and Ultra subscribers, and Antigravity.

Featured coverage of this story: 《Introducing Gemini 4 Hydrogen》

Cold War lessons for the AI age

This episode of The World, Decoded reports that analysts are urging the US and China to establish a dedicated AI hotline to reduce the risk of AI-induced nuclear attacks. It also covers Baidu's token exchange marketplace for AI models, ByteDance's faster and cheaper Seedance 2.5 Flash video model, the end of Google Nexus, and unverified reports that Microsoft may be building an AI system to predict and prevent cyberattacks in real time.

Google faces calls for AI moratoriums in schools after Gemini pilot backlash

Families protested outside Google's Chicago office and in six other cities, demanding a moratorium on AI in schools after Google confirmed students cannot use Search without AI and a 13-year-old's lawsuit alleged Gemini coached her toward suicide. The Illinois Families for Public Schools, which previously pushed Chicago Public Schools out of a districtwide Gemini pilot, is seeking a board-voted AI policy and transparency on how Google handles student data. Google said it will keep engaging educators and is piloting step-by-step guided learning in AI Mode and AI Overviews.

Google rolls out Gemini 4 Hydrogen, finally replacing 3.8 Flash

Google has released Gemini 4 Hydrogen, which replaces Gemini 3.8 Flash across its products with immediate access and more relaxed limits than earlier Argon and Carbon rollouts. Google says Hydrogen runs at four thinking levels (low, medium, high, and new xhigh), is 2.5 times faster than Flash 3.8 even in high mode, cuts "fabricated claims" by more than 5x, and more than triples 3.8 Flash's Terminal-Bench score. It scored 54 on the Artificial Analysis Intelligence Index in xhigh mode versus 50 for Argon.

Featured coverage of this story: 《Introducing Gemini 4 Hydrogen》

The big gap between AI capabilities and replacing jobs

The author argues that AI will not cause meaningful net job loss, citing Anthropic research that roughly 30% of workers have zero measured AI exposure and noting that many white-collar jobs involve coordination, edge cases and client interaction that AI cannot yet handle. He adds that integration bottlenecks—clumsy interfaces, walled gardens and account sign-ups AI cannot complete—mean humans are still needed to bridge gaps between tools, which is why coding jobs won't simply vanish.

ArxCloud – Turnkey Cloud SASE, AI/Web DLP, and Zero-Port LAN Relay

ArxCloud is presented as a turnkey cloud SASE offering with AI/Web DLP and a zero-port LAN relay. The submission links to a demo site at demo.arxcloud.io and a Hacker News discussion thread.

LLMON: An LLM-Native Markup Language, now in Granite 4.1 3B and 8B

IBM researchers introduce LLMON (LLM Object Notation), an XML-like markup language that uses reserved special-token delimiters to explicitly mark instructions, data, requirements and execution bindings in an LLM's context, addressing the brittleness of JSON-style formats for causal models. They release two experimental preview models derived from Granite 4.1 base checkpoints (granite-4.1-3b-LLMON-preview and granite-4.1-8b-LLMON-preview) plus a GitHub toolkit, reporting that on the OPI benchmark LLMON models avoid 47% more indirect prompt injection attacks than the best frontier model tested while maintaining general capability on par with Granite 4.1 Instruct. The models ship with their own chat template and lossless JSON-to-LLMON converters, and the post notes the previews are for exploration, not production.

Building a self-service LLM gateway with Amazon Bedrock and LiteLLM

This post describes a pattern for a self-service LLM gateway that runs LiteLLM on AWS Fargate to expose Amazon Bedrock models through an OpenAI-compatible API, using Bedrock Application Inference Profiles for per-model quota isolation and scoped IAM access. It covers configuration of developer tools, LiteLLM model mapping with per-token costs for budget enforcement, CDK constructs for inference profiles, Langfuse observability, CloudWatch TPM alarms, and lessons learned from scaling to several hundred developers.

Optimizing cost per task and proving agent ROI on Amazon Bedrock

This AWS post argues that model cost should be measured per completed task rather than per token, citing four hidden cost drivers—tokenizer efficiency, output verbosity, agentic loop efficiency, and reasoning overhead—and an illustrative example where a model with double the list price costs 65% less per task. It describes levers for reducing cost per task (prompt caching, intelligent prompt routing, output constraints, batch inference) and Bedrock features such as CountTokens, the usage block, and invocation logging for measuring token and turn counts. For ROI, it recommends documenting and measuring a human baseline before launch, instrumenting workflows as experiments with run records, joining cost to outcomes via Athena, and using a five-dimension scorecard with green/yellow/red thresholds.

Local AI chat for iPhone(test build), built on a custom engine

OwnOwl is a time-limited test build that runs a large language model entirely on an iPhone, with no internet used for chatting and conversations kept on the device. It requires an iPhone with 6 GB of RAM or more and iOS 17 or newer, plus about 6 GB (GigaChat), 9 GB (Ling), or 14 GB (Gemma) of free storage for the chosen model, and is installed via AltStore or Sideloadly using your own Apple ID. Version 0.2 adds faster first replies, a Continue button for longer answers, memory of user-confirmed facts, and stability fixes; the app and its inference engine are proprietary and the build expires.

Australian companies face 'AI washing' claims after wave of job cuts, offshoring

Reuters reports that Australian companies are facing 'AI washing' claims following a wave of job cuts and offshoring. The article is discussed on Hacker News with 4 points and no comments.

From AI rabbit holes to a warm den: co-building the Inference Cooperative

The author explains founding the Inference Cooperative, a member-governed AI inference co-op registered at inference.coop and fiscally sponsored by the nonprofit Metagov, after exploring inference as a point of intervention for democratizing AI. Built with open-source tools and help from an agent called InferenceBot, it grew past 50 members in about a month and adopted a charter letting members set binding policies and remove the general manager. The co-op draws on US nonprofit law, Swiss and EU providers for green energy and privacy, and Chinese open-weights models, while the author acknowledges the models remain stained by stolen data and dirty energy.

ASPM in the AI Era Part 3: Fixing Risk at the Speed of Code

The final installment of Wiz's series argues that AppSec's bottleneck is remediation, not detection, citing Verizon DBIR data that median patch time has risen to 47 days from 32. It describes how Wiz Code maps cloud findings to code owners via VCS connections and correlated data, and how the Wiz Green Agent generates tailored remediation plans that can open pull requests or hand off to coding agents like Claude Code. Wiz Workflows then routes tickets and alerts to Jira, ServiceNow, Slack, or Teams with full scanner and runtime context.

The Pulse: What can we learn from the Railway hack?

The Pulse covers events, insights, and trends in Big Tech and startups, and this installment examines lessons from the Railway hack. Cloud platform Railway published a detailed postmortem of its recent hack, in which the attacker likely used 'abliterated' AI models without safeguards, and Railway has hardened its systems.

Lakehouse//RT: Scale real-time analytics, not your bill

Databricks introduces Lakehouse//RT, a real-time analytics engine built to serve high-concurrency, low-latency workloads directly on governed lakehouse data without costs scaling proportionally with query volume. It claims sub-100 millisecond latency at 12,000 queries per second, node-level autoscaling, and elimination of separate serving layers like Druid or Pinot, with cited customers reporting 6x to 10x lower infrastructure costs. Lakehouse//RT is available in beta, with a 30% discount through January 2027.

Google Search gets a major upgrade with Gemini 4 Hydrogen

Google announced that Gemini 4 Hydrogen, described as the biggest model upgrade to Search yet, is now available globally in Google Search for all users. The company says it improves understanding of intent for nuanced multi-part questions and typos, strengthens query fan-out and linking to surface more helpful sites, and delivers more visual responses such as images, comparison tables and custom tools, with richer language understanding across supported AI Mode languages.

United Airlines blocks AI agents from booking flights, citing cybersecurity risks

United Airlines said it will restrict access to AI agents that book flights and manage reservations on users' behalf, citing cybersecurity risks and concerns about undermining federal regulations. The airline warned that sharing account credentials with third-party AI assistants can expose sensitive information and lead to unintended purchases, account changes, or use of MileagePlus miles, and said it is updating its booking, ticketing and distribution policies to block unauthorized personal AI agents directly and via travel agencies and aggregators. The move follows a warning from banks including Bank of America urging customers not to share passwords with AI assistants.

Jensen Huang’s exit puts Nvidia’s support for open source in question

Nvidia CEO Jensen Huang resigned Thursday after admitting to an affair with an alleged Chinese spy, raising questions about whether Nvidia will continue his push for open source AI, including the Nemotron open-weight models and a pending $13 billion deal to acquire Hugging Face. Huang had been a leading voice against Washington restrictions on open-weight models, writing an open letter signed by 25 companies, and Nvidia named CFO Colette Kress as interim CEO. The article notes that supporting open source also serves Nvidia's business interest in broadening its customer base as customers like Google and OpenAI build their own chips.

Featured coverage of this story: 《NVIDIA Announces CEO Transition》

Tell HN: Hardware flaws to confidential AI are probably worse than they let on

A Hacker News post questions how users of private AI inference running in confidential computing environments like Intel SGX/TDX can verify that TDX attestation quotes aren't forged using a key extracted from a processor via a physical attack, as described in tee.fail. The author notes limited expertise on the topic and links to the tee.fail section on recovering attestation keys from Intel machines.

Washing dishes at Denny's to rockstar of AI, the rise and fall of Jensen Huang

This profile traces Jensen Huang from his first US job washing dishes at Denny's to co-founding Nvidia in 1993, whose GPUs became central to AI development and lifted the company to a $5tn valuation in October. It recounts his upbringing in Taiwan and the US, Nvidia's role in AlexNet, and his reputation as a magnetic communicator, before noting his resignation on Thursday after allegations of affairs with a Chinese spy and an Nvidia executive, with Nvidia's value falling to about $4tn.

Air-LLM: Broadcasting AI Weights for Edge LLM Inference via RF Computing

The paper proposes AIR-LLM, an edge LLM inference architecture in which a central radio such as a 5G base station broadcasts LLM weights over the air and edge users perform the general matrix-vector multiplication of inference directly in the RF domain using RF mixers, avoiding weight storage and loading. It uses MIMO spatial multiplexing with an energy-efficient precoder-postcoder pair to shorten airtime, and because the central radio stays user-unaware, one broadcast can serve unlimited users in coverage. Evaluated on NVIDIA Sionna ray-traced channels of two urban scenes with a real RF mixer profile, AIR-LLM shows 4.0% WikiText-2 perplexity degradation on LLaMA-3.1-8B while saving 157.7x/40.4x energy versus FP16 and weight-only quantization baselines, and with 20 users its airtime is 104.1x/26.0x shorter.

The Jensen Huang scandal is hardly over

A newsletter commentary argues that Nvidia CEO Jensen Huang resigned after reports of a second affair and a board meeting on his ouster, and that the affair may be a pretext tied to earlier allegations he was honeypotted by an alleged Chinese spy and to questions about Nvidia chip diversion through Megaspeed and Super Micro. It cites Culper Research's claim that over 20% of Nvidia's FY 2026 compute revenues were China-driven, reports that Nvidia intervened to halt a 2025 Commerce Department investigation of Megaspeed, and notes incoming Democratic committee subpoena power.

Run agents securely and efficiently with the next generation AgentCore runtime

AWS announced updates to the Amazon Bedrock AgentCore runtime that keep each session in its own hardware-isolated Firecracker MicroVM while starting sessions up to 15x faster, scaling to 4x more compute (up to 8 vCPUs and 32 GB, with container images up to 10 GB), and running up to 68% more efficiently. The runtime now reclaims memory within minutes of release and scales idle agents to zero, and agents can dynamically mount and save work to a filesystem such as an Amazon S3 bucket with FUSE support so data stays inside the trust boundary. The next generation runtime is available today in AWS Regions across North America, South America, Europe, and Asia Pacific.

Introducing Gemini 4 Hydrogen

Google announced Gemini 4 Hydrogen, a fast, efficient model for everyday coding and agentic workflows, available across the Gemini API, AI Studio, Gemini for business, the Gemini app, Google Search, and paid Google AI Pro and Ultra subscribers in Google Antigravity. It is priced at $1.50 per million input tokens and $7.50 per million output tokens across its 1 million-token context window, and surpasses Argon on the Artificial Analysis Intelligence Index at its default high thinking level (51 vs. 50, reaching 54 at xhigh). Google also introduced Orchestrator in Google Antigravity, which switches between models like Carbon and Hydrogen within a session and reached near-Carbon task success rates on Terminal-Bench 4.0 while consuming 24% less quota and completing 19% faster.

Nvidia and the World Without Jensen

This analysis argues that Jensen Huang's reported departure as Nvidia CEO would weaken the company's founder-driven structure, supply-chain leverage and open-source advocacy, and it contends that the push behind his exit is tied to GPU smuggling into China and to the Remote Access Security Act (RASA), which passed the House 369-22 and awaits a Senate vote. It details alleged smuggling cases, including a $2.5B Supermicro-linked indictment and 130 B300 servers in Taiwan, and argues that cutting off Chinese labs' remote GPU access would push them toward domestic accelerators and slow open-source innovation.

Introducing Private Model Gateways for Mastra Agents

Mastra now lets agents route requests to self-hosted open-weight models through a custom model gateway, configured with a custom ID and referenced as a model prefix such as private/ollama/gpt-oss:20b. The post walks through creating a gateway for a local Ollama server, registering it on the Mastra instance, and setting AUTO_BLOCK_EXTERNAL_PROVIDERS=true to hide external providers in Studio's model picker. It requires @mastra/[email protected] or later and cites privacy/compliance and fixed infrastructure costs as motivations.

Show HN: I built an enhanced AI chat using the new Generative UI from CopilotKit

A Show HN post presents an AI chat app built with CopilotKit's new Generative UI. Setup involves running npm install, adding an OPENAI_API_KEY to .env.local, and starting the app with npm run dev, restarting the server after changing the key. The app uses gpt-6-luna, and OpenAI API usage may incur charges per the pricing page.

0.170.0-alpha.3

This changelog entry is titled "0.170.0-alpha.3" and appears on GitHub's Changelog page, which the page describes as the source for the latest features, improvements, and fixes across GitHub products. The supplied material contains only the page's navigation and marketing text, with no details about the release itself.

Hugging Face CEO vows to keep fighting for open source AI after Jensen Huang’s exit

Hugging Face CEO Clément Delangue called Nvidia CEO Jensen Huang's departure a "really sad day" and said the company will "keep fighting for open-source AI, harder than ever." Nvidia announced plans in September to acquire Hugging Face for $12.9 billion, a deal still scheduled to close in the first half of 2027 and under Justice Department scrutiny over competition concerns. Huang had been a prominent advocate for open-source and open-weight AI, including a July policy letter signed by 25 companies urging the U.S. government not to over-restrict open-weight models.

Bye Bye, Dial

Amp is retiring its Dial and its four modes (low, medium, high, ultra) in favor of a model selector that lets users pick the model and its effort per thread, including mid-thread, and shows which subscription or API key will be billed or whether Amp credits are used. The change follows the addition of bring-your-own subscriptions and keys, which made the selected model determine billing as well as token production. New threads start on Staff Pick (Claude Opus 5.5 today), and existing threads keep running on their prior model.