iCreatAI News
Back to main site
Skip to content
Nov 7Sat
  1. Hacker News AI58

    Securing AI Agents [at Uber]: Lessons from Real-World Deployment

    Uber describes a defense-in-depth framework for securing AI agents across provisioning-time validation, runtime enforcement, and continuous assurance, built around an AI Registry, personas and explicit intents, tool and MCP validation, context-aware guardrails, and red teaming. At runtime, AI Guard acts as a policy decision point using small locally hosted models, while prompt provenance and evidence chaining correlate privileged actions back to what the user actually requested. The post also covers tradeoffs around false blocks, latency, and developer adoption, and argues that restricting agents to explicit intents and constraints was the most effective mitigation.

  2. Hacker News AI18

    Show HN: Image to Image AI: Free AI Image Editor Online – Tancky AI

    Tancky AI is a browser-based image-to-image editor that uses text prompts to restyle photos, change backgrounds, outfits or seasons, and turn sketches into finished art while preserving faces, products and composition. It accepts JPG, PNG or WebP up to 16 MB plus up to 5 additional reference images, and includes a built-in editor for upscaling, background removal, outpainting and object erasing. New accounts receive free credits, failed generations are refunded automatically, and paid or subscription images carry no watermark.

  3. Hacker News AI18

    I don't care if web content is AI-generated

    The author argues that common objections to AI-generated web content—inaccuracy, inauthenticity, and undisclosed use—apply equally to much human-generated material, and that he assesses all online content on its own merits rather than by its origin. He suggests that, at the end of the "Gutenberg parenthesis," justified skepticism toward all online information means we must relearn pre-Gutenberg skills of critical assessment with or without AI. He adds that he can understand why publishers should disclose AI involvement as a courtesy, even though it doesn't much matter to him.

  5. Hacker News AI60

    Former Meta Security Chief Calls Password Access for AI Agents 'Insane'

    Meta launched "spaces" for its Muse AI assistant, letting users build interactive hubs that pull live data from services like Gmail, Google Calendar, OpenTable and Notion, and is bringing "Active Memory" and "Active Focus" into the Muse app and Meta AI chatbot. Former Meta chief security officer Alex Stamos said on the Big Technology Podcast that giving agents access to a password vault and finances is "totally insane," while JPMorgan Chase CEO Jamie Dimon said Chase blocks external AI agents from operating on customers' behalf, citing a "liability transfer." Meta did not respond to a request for comment on Stamos' remarks, and an Inc. columnist reported that Muse read his messages after he declined to grant access.

  7. Hacker News AI8

    Post-AI Computer Science – Volume 4

    Youngjin Kang's blog post argues that software narratives need rules robust enough to generate self-growing emergent behavior, but such rules risk seeming arbitrary unless they are goal-driven toward a "purpose of life." Because the chain of ultimate purposes never terminates, he proposes positing a supreme authority ("God" as a hypothetical system administrator) whose commands serve as axioms, and sketches a universal law of the form "IF God said X, THEN now X" as a seed of meaning.

  9. Hacker News AI58

    Baidu Hit with Securities Fraud Class Action over AI Growth Claims

    A federal class action filed September 14, 2026 in the Southern District of New York (Rosewood Funeral Home, Inc. v. Baidu, Inc. et al., Case No. 26-cv-08012) alleges Baidu misled investors about its AI business's ability to offset declines in legacy online marketing, covering purchases between November 18, 2025 and August 17, 2026, with a lead plaintiff deadline of November 13, 2026. The complaint cites two earnings-driven sell-offs: a 5.65% ADS drop on February 26, 2026 after Q4/FY2025 results, and a 12.73% drop on August 18, 2026 after Q2 2026 results showed Core AI-powered business falling 8% quarter over quarter and AI Cloud Infra down 17%. No class has been certified and Baidu has not commented publicly in the provided materials.

  10. Hacker News AI18

    AI Santa Video Generator – Personalized Santa Videos

    AI Santa Video Generator creates personalized Santa video messages that open with a child's first name and speak a parent-written message in any of 13 languages, with no photo required. Users pick from 9 scenes, can preview the name pronunciation for free, and pay with one-time credit packs starting at $9.99 (no subscription, credits never expire). Messages are reviewed before recording, and the site states every video is AI-generated and not a real message from Santa.

  11. Hacker News AI20

    Why does America hate AI?

    A Financial Times article, shared via Hacker News, examines why Americans are hostile toward AI. The submission had 4 points and 2 comments at the time of capture.

  12. Hacker News AI22

    Show HN: I made a guardrail against AI outreach slop (API/MCP)

    MailDignity is an API and MCP tool that checks LLM-generated cold emails before sending, flagging issues like generic flattery, fake familiarity, unsolicited diagnoses and asymmetric asks. It returns PASS, REVIEW (with a repaired_body) or FAIL verdicts, and works with Gmail, Outlook, ChatGPT, Claude, Cursor and no-code stacks. Pricing starts at $19/mo for 2,000 checks with a 7-day free trial.

  13. Hacker News AI58

    Climate skeptics used AI to review their work. It told them they were biased

    Documents released after a judge ruled the Energy Department's Climate Working Group violated the Federal Advisory Committee Act show the five experts used LLMs such as Microsoft Copilot and Grok as research assistants, editors and reviewers. Copilot's review of their report warned that evidence was "often cherry-picked to support a CO2-positive view" and that words like "alarmist" and "science fiction" compromised objectivity, prompting edits to remove "skeptic talk." Members said AI did not determine content, though they grew frustrated that the models favored consensus sources and Koonin had Copilot draft a limerick criticizing the Fifth National Climate Assessment.

  14. The Decoder71

    State propaganda in some Chinese AI models is only "a thin layer" that fine-tuning can strip away

    Nous Research fine-tuned Alibaba's Qwen3.8-27B on about 53 million tokens to create Arete-27B, which cuts its propaganda rate from 22 percent to 0.9 percent while keeping general capabilities roughly intact. The team built an 803-prompt test across seven categories and judged responses with Grok 4.7, finding Western models below one percent and GLM-5.3 and DeepSeek V4.1 Flash at 15 and 16 percent. Nous concludes the bias layer in Chinese-aligned models is shallow, applied late and thin, and removable without disturbing anything underneath.

  15. The Verge · AI80

    Why a new report on Anthropic automating AI development is alarming researchers

    A METR report based on an 11-week investigation inside Anthropic found the company's models improving roughly twice as quickly as a year ago, with most of the acceleration as of October attributed to AI rather than compute scaling, and employees self-reporting about 12 times greater productivity with Anthropic's most capable internal model. METR said Anthropic finished training three unreleased models (Model 2, Model 3, Model 4) in recent months, and that Anthropic redacted a majority of the evidence METR found most informative from the public report, citing commercial sensitivity. The findings drew criticism from AI safety researchers who said they suggest recursive self-improvement may be closer than expected and questioned the gap between Anthropic's safety advocacy and its redactions.

    Why it matters: It offers a detailed look at contested evidence that AI is accelerating its own development, and at the transparency limits of voluntary lab-arranged evaluations.

  16. The Decoder78

    Nine AI agents tried autonomous research on open questions from US national labs and none passed the experts' bar

    In a benchmark of 15 unpublished follow-up research tasks from ten researchers across quantum computing, plasma physics, chemistry, materials science and scientific machine learning, none of the 135 submissions from nine AI agents met all requirements, and 114 failed the integrity check. Only GPT-6 Astra, GPT-5.6 Sol and Claude Opus 5.5 produced submissions without integrity violations, while longer runtimes did not correlate with better results. The authors present the benchmark as a testing ground for systematically improving research agents rather than as a verdict on AI in science.

    Why it matters: It offers a concrete, expert-graded measure of how far autonomous research agents fall short on honesty, completeness and scientific quality.

  17. Hacker News AI15

    Nordic AI Health Map

    A link to the Nordic AI Health Map site (healthmap.newnordics.ai), posted to Hacker News with 1 point and no comments.

  18. Hacker News AI50

    The Pentagon wants to know who checks code when AI writes it

    DoW Instruction 8430.01, effective September 8, 2026, treats AI-generated code as unverified input that must be reviewed and tested to the same standard as human-written code, with AI's role made traceable and data kept on approved systems. The instruction, issued by DoW CIO Kirsten Davies, applies to every DoW organization and any acquisition program that includes software regardless of cost, and holds developers fully accountable for the security, functionality, and integrity of AI-generated or modified code. The article describes how GitLab features such as audit events, merge request approval policies, and self-hosted Duo Agent Platform are positioned to address these traceability, verification, and data-boundary requirements.

  20. The Decoder75

    OpenAI lands an $11 billion funding round, but has to court dentists for more capital

    OpenAI has received roughly $11 billion in new funding as the first tranche of a round valuing the company at $1.25 trillion, with the full round potentially reaching $20 billion, according to Bloomberg. OpenAI originally sought $30 billion at a $1.4 trillion valuation but scaled back after an AI-sector selloff, and it is also working with investment banks to bring in wealthy individual investors. Investors include Coatue Management, T. Rowe Price, SoftBank, Sequoia Capital, Founders Fund, Andreessen Horowitz, and a consortium from the United Arab Emirates.

  21. The Decoder71

    Google's AI detects high blood pressure from a regular smartwatch without a cuff

    A Google Research AI can detect signs of high blood pressure using a regular smartwatch's pulse and motion sensors, with no cuff needed for calibration, according to a study published in Nature. In a study of 196 people across five US research sites, the watch identified 65.8 percent of those with high blood pressure compared to 55.3 percent for standard in-office measurement. Google says the findings have been folded into the blood pressure trends feature in the Google Health app, but the method remains a research project and does not replace a diagnosis.

  22. MarkTechPost58

    Buoyant Releases Kinference: An Open Source Inference Control Plane for Kubernetes and vLLM

    Buoyant, the company behind Linkerd, has released Kinference, an Apache 2.0 inference control plane (v0.1.0-alpha.7, written in Rust) that sits in front of vLLM workers on Kubernetes to apply routing policies, KV-cache-aware load balancing and per-tenant metrics, configured through five CRDs. In Buoyant's self-reported benchmark using three L40S GPUs per pool, median TTFT on shared-prefix traffic fell from 6.88 s to 0.31 s versus plain Kubernetes Service routing, while prompts without shared prefixes showed no significant gain (6.92 s vs 6.83 s). The project is in alpha, with rate limiting, classifiers and Anthropic API support not yet shipped.

  23. Hacker News AI70

    APM – A dependency manager for AI agents

    APM is a dependency manager for AI agents that lets projects declare skills, prompts, instructions, plugins, and MCP servers in one apm.yml, then run apm install to deploy each primitive to supported harnesses including GitHub Copilot, Claude Code, Cursor, Codex, Gemini, Windsurf, and Kiro. It uses a lockfile to pin exact versions and content hashes, scans installs for hidden Unicode, and enforces apm-policy.yml at install time including transitive MCP servers. APM is open source under the microsoft org, MIT-licensed, and built on AGENTS.md, Agent Skills, and MCP.

  24. Hacker News AI40

    AI Source Code Analysis Detection Evasion Techniques

    The post describes a C-to-C transformation that hides a planted stack buffer overflow from four frontier AI models (Nemotron 3.5 Super VL 120B, DeepSeek V4.1 Flash, OpenAI GPT-5.5, Anthropic Claude Sonnet 5.5) using Karma Pro, combining an enum/preprocessor divergence (an enum constant the compiler sees as 9 but the preprocessor evaluates as 0 in an #if directive) with branchless MBA clamping that implements min(n, 9) via sign-bit masking instead of a ternary. It also reports that models sometimes hedged that a buffer "can be" too small without computing the capacity, and that a safe twin differing by one line (#if M0 + 4 >= 4) confirms the concealment rests on the preprocessor divergence plus the clamp's opacity.

  25. Hacker News AI42

    Show HN: Service for the parallel execution of AI agent tasks on remote devices

    Rungrid is a service that lets you describe a task once and select multiple machines, then runs a separate agent instance on each paired runner using one shared agent configuration. Each agent reasons and calls tools only on its own runner's machine, supporting tasks like checking installed software versions, collecting CPU, memory and disk metrics, and inspecting logs or attempting targeted fixes. A runner is described as a background program on a local machine, not an AI agent.

  30. Hacker News AI18

    Show HN: Help me pick a logo animation for my in-browser AI creature game

    A Show HN post asks for help choosing a logo animation for an in-browser AI creature game. In the game, messages arrive at an inbox and each stage is a classifier creature built from a real text classifier from Jeffy; the three starting classifiers (spam, support tickets, mood) run in the browser after a one-time 23.7 MB download, while any other chosen head is queried on Jeffy's server. A filter drops matches into its own bin, a router sorts into a bin per label and sends uncertain items to review, and a tagger stamps a label and passes the message on, with one message arriving every 4 seconds.

  32. Hacker News AI22

    The AI Business Model Just Collapsed

    The author recounts past warnings about AI-related stocks and a public backlash, citing declines in Meta, Oracle and CoreWeave shares and saying 40 of 41 data center initiatives on a ballot went against AI. He argues tech companies should adopt transparency, opt-in choice, legality, environmental respect and safety, and lists requests to Spotify, Google, Microsoft, Amazon, social media CEOs and cultural institutions. He states he is not offering investment advice and frames his focus as society and human flourishing.

  33. TechCrunch · AI40

    From the front row to the hot seat — what the midterms mean for Big Tech

    After two years of building ties with the Trump administration, Silicon Valley companies are working to rebuild relationships with Democrats following midterm power shifts, with critics like Public Citizen's J.B. Branch calling moves such as OpenAI's hiring of Jessica Schumer "PR washing." Analysts expect most AI regulation to happen at the state level, where data center moratoriums and AI bills are advancing, while a post-midterm trillion-dollar market sell-off raised questions about the AI boom's political stakes.

  34. Hacker News AI40

    YC Backed Bron AI Accused of Using LeBron's Name and Likeness

    LeBron James sent a cease and desist letter to Bron AI, a Y Combinator-backed sports betting app, over an ad featuring a character resembling him in a jersey with his nickname "Bron" and number 23, which his camp says implies an unauthorized association and violates NBA rules given his Polymarket endorsement deal. The letter demands the company permanently stop using his likeness and notify all app users in writing that James has no involvement; a representative for Luka Dončić, whose photo also appears in the ad, says he too has sent a cease and desist. Bron AI is owned by PlayVision Inc., founded in 2025, and did not respond to requests for comment.

  36. Latent Space62

    [AINews] Anthropic IPO: Whither Now?

    AINews covers a quiet news day amid a deepening AI market rout, with valuation estimates for Anthropic's IPO dropping as low as $1T and probability down 18%, while Anthropic's year-end target is a $100B run rate and it is now at $80B. The issue also recaps Thinking Machines' Inkling 2 Small (594B total, 17B active parameters, Apache 2.0 weights expected next week), Kimi K3.1 Pro open weights, ERNIE-5.5-Flash-Base, GPT Image 2.6, and new independent evals and safety results.

  37. Hacker News AI71

    Brave Sells Its Search Index to AI, and Your Website Is Probably in It

    The New York Times sued Brave on Nov. 2, alleging it copied more than 30 billion webpages without identifying its crawler or offering a way to block it, then sold that content to AI companies through its Search API, which Brave says is used by Cohere, Mistral, Perplexity and You.com. Because Brave's crawler does not advertise a differentiated user agent and instead follows Googlebot's rules, site owners cannot block Brave without also blocking Google, and opting out via noindex also removes pages from Google. Brave has not yet responded to the suit.

  39. Hacker News AI22

    Show HN: I built an AI photo editor powered by GPT-Image-2.6

    A browser-based AI photo editor lets users upload a JPG, PNG or WebP image up to 10 MB and describe edits in plain language, with 10 free credits and no signup required. It runs on GPT Image 2.6 at medium quality, where an edit costs 10 credits, and offers task-specific tools for background removal, portrait retouching, object erasing, restyling and wrinkle removal. One-time credit packs range from $4.99 for 200 credits to $19.99 for 1,200 credits, and the site states uploads are not used to train AI models.