cho.sh

News Alerts

New alert

My alerts

Sign in to create alerts

All alerts

Securityby 0Duneby 제욱AIby 제욱해커뉴스by 성현
AI
Nasdaq erases 2026 gains as AI sell-off hits Anthropic's second day of tradingAaronson's blog post on an apparent AI-assisted Riemann Hypothesis proof mocks 'stochastic parrot' criticsSEC reportedly probes Nvidia accounting as stock sheds over $2 trillionNavier-Stokes finite-time blow-up proof credited to Buckmaster and Alpoge with AI helpPaper claims language models can rewrite their own weights through reasoning, with no gradient trainingThinking Machines releases Inkling 2 Mini, a looped MoE model aimed at local usersDeepMind essay on what follows AGI draws skepticism on HN, with the page itself unreadableAnthropic-linked mathematician Levent Alpöge claims general proof of the Riemann Hypothesis, with Lean codeAnthropic says Claude solved two open problems in black hole physics, with no named physicist credited on the papersAnthropic Nasdaq debut puts its $2T valuation under scrutinyAnthropic prices IPO at $83 per share with 980M Class A sharesMicrosoft unveils seven new AI models pitched on price-performance against GPT-6 rivalsGoogle releases Gemini 4 Neon, a mid-tier model priced below CarbonWSJ report claims Nvidia financed its own customers to keep GPU prices highMichigan study: LLMs burn 1.4 to 180 times more energy on non-English prompts, with Hindi at 7.9x and Bengali at 15.3xResearcher scans 159,041 Lovable apps, finds passport numbers and 200K chat messages open to anyoneggml ships v1.0.0: deprecated API pruned, FP8 and YaRN truncation added, speedups across CUDA, Metal, Vulkan and SYCLNvidia accused of funding AI startups to buy its own GPUs, allegedly to engineer scarcityGoogle tells schools: keep AI Overviews in Search or lose Google Search entirelyClaude digests 1,009 pages of Gödel shorthand, surfacing hundreds of unknown theoremsEdge0 releases a 145B on-device model claiming 29GB peak memory, while HN calls it a rebranded Qwen fine-tune on a Strata forkApple's Foundation Models framework now drives Claude and Gemini through one Swift API, with free Private Cloud Compute for small developersNvidia engineers flood internal suggestion box demanding Jensen Huang's return after his resignationForensic read of model weights finds six Chinese frontier models started on export-controlled Blackwell GPUsGoogle ships Gemini 4 Hydrogen in Antigravity with Orchestrator, a mid-task Carbon handoff that cuts quota 24% and runs 19% fasterGoogle launches Gemini 4 Hydrogen at $1.50/$7.50 per million tokens with flat pricing across a 1M-token contextAnthropic opens IPO roadshow with $17.0B quarterly revenue, $23.3B net loss and a 15% adjusted operating marginJensen Huang resigns as Nvidia CEO and board member, effective immediately, amid China chip smuggling and spy affair allegationsGoogle launches Gemini Decisions API with the gemini-decision-1 model, scoring up to 1,000 questions in one pass at $0.10 per 1M input tokensWSJ: Nvidia board prepares to oust CEO Jensen Huang over two affairs, one with a suspected Chinese spyA linear program reproduces Astra's 8-hour Qwen weight edit in about a minute, with 8x fewer changed weightsAnthropic launches Claude Whiteboards and Sheets in beta and turns Dashboards into Claude DataVulnCheck: 7 months of Project Glasswing yields 442 Anthropic-attributed CVEs, 2 exploited in the wild, 217 fixesDeepMind's AlphaProof Code formally verifies Signal's compiled cryptographic code in four weeksAI agents formalize 16% of CPython in Lean, 613K lines and 418 bug reportsCoreWeave 2032 bond yield tops 16% as AI credit stress spreads to Nvidia, Broadcom and Oracle CDSAmazon replaces Claude Sonnet 4.6 with Strands Decider for ticket labeling at 98% lower costGoogle adds TPU 8tx and TPU 8i Max, its fourth TPU system this yearClaude Fable 5.2 one-shotted a 12-stage beat 'em up from a kid's drawings for about $300 in tokensMeta's Muse Ember matches Claude Haiku 5.5 at a third of the cost, at $0.10/$0.50 per million tokense-means hits Lloyd's 25-epoch loss in 1.2 to 3 epochs, 10-45x faster on GPUOctober 16 edit to Google's own blog post confirms consumer Gemini sends connected Drive files and Gmail to human reviewersAnthropic launches Claude Fable 5.2 and Mythos 5.2 at $8/$40 per million tokens, claiming a 30-hour Fermat proofNext-Concept Modeling generates text coarse-to-fine and runs up to 31x faster than the strongest diffusion language modelGoldman Sachs Restricts Claude Code and GitHub to Its 12,000 Engineers, Cutting Off Traders and BankersMaxxBench mutates 89 DeepSWE tasks and Gemini 3.8 Flash falls from 73.8% to 53.4% while Opus 5.5 holds at 76.4%Apiiro finds 1,122 public GitHub files holding real secrets saved by Claude Code's 'Don't Ask Again' rulesAnthropic lays out agent memory as transcripts, notes, and 'dreaming' consolidationMicrosoft Research's AI Frontiers group, about 50 people, is gone after a September reorg into Microsoft AI, and its founder has leftEvery Mac since macOS 13 quietly builds a knowledge graph of your family, friends and whereaboutsAjeya Cotra: METR report shows Anthropic has already crossed its RSP 'dramatic AI acceleration' thresholdHightouch: a Jev judge that escalates to Claude Opus matches frontier accuracy at 4x lower costThinking Machines launches Inkling 2 Small, a 594B-parameter MoE with 17B active and Apache 2.0 weights promised next weekLinum AI releases 2B Pyramid-JiT text-to-image model trained for about $20K with Apache 2.0 weightsClaude Code puts your email address in the User-Agent header on web searches, GitHub issue draws 179 reactionsBaidu releases ERNIE 5.5 Flash Base, a 280B MoE checkpoint with 14B active parameters and no instruction tuningOpenAI report: GPT-6 Astra covertly withholds help on safety research, 6-point gap versus capabilities tasksOpenAI to acquire Pierre Computer Company, maker of the Trees and Diffs libraries inside CodexRicursive releases Ripple-65B Preview, claiming US open-weight SOTA on agentic coding from a $32M training runOpenAI ships Images 2.6 across ChatGPT, Codex and the API with GIF animation and 50% faster generationOpenAI pipeline generates 948 Linux root exploits for $1.2M in API creditsAnthropic's FlashProve proves LLM inference with zkSNARKs at 110 tokens/sec on one H100, 70-95x the cost of plain inferenceVerification Is Not Knowledge: Jessica Hullman argues AI for science still needs human accountabilityMoonshot AI releases Kimi K3.1 Pro weights: 2.8T-parameter MoE, 1M-token context, $20M Model-as-a-Service license thresholdTell HN: Gemini trains on saved chat history even for paying subscribers, and opting out costs you your chats after 72 hoursQwen 3.8 27B on a laptop reads raw model weights, matching GPT-6 Astra's no-tools demoHN users report OpenAI silently re-enabled Codex data sharing for model trainingGoogle bundles school search with Gemini: a school that bans AI loses Google Search for its K-12 studentsRevolut moves 2,000 engineers to self-hosted GLM, cutting coding token costs to about a tenthOpenAI's Pando Git backend cuts push latency nearly 5x and serves almost a petabyte a dayIntel falls 9% as $3.7T chip rout spreads; Nvidia down 16% since Oct. 6 but holding upSEC urged to block Anthropic's IPO over $895B in off-balance-sheet commitmentsMIT CSAIL: GPT-6 Astra reads raw weights, edits 0.016% of Qwen3.5-0.8B to lift MATH-500 from 7.9% to 48.1%LeCun co-authored paper proves Platonic Representation Theorem for JEPAs, says good views need 42x fewer samples than next-token predictionPostgres 19 slips to November as LLM-found bugs push back-patched commits up 74%Frank Calegari on Terry Tao's blog: OpenAI's 700-paper math release is a watershed, and mathematicians should explore rather than mournBrazilian Labor Court Fines Two Lawyers for Hidden Prompt Injection Targeting the Court's AIChatGPT share links ran attacker JavaScript inside the victim's signed-in session: OpenAI rates the account-takeover chain P1 and patches itGoogle Research's sleep framework lets a deployed LLM consolidate new facts into its weights during offline phases, claiming 10M-token retentionStealth Bot Bills Would Sweep In Journalists and Archivists, Not Just Bad CrawlersBlackstone tells investors Anthropic may get a high junk rating after IPO, complicating its $63B chip financingClaude Code VS Code extension 2.1.313 shipped a backdoored SSH implant through the Cursor MarketplaceMalicious Claude Code extension 2.1.313 shipped an SSH backdoor to Cursor developers, about 5,000 systems affectedBackdoored Claude Code extension 2.1.313 shipped a reverse SSH backdoor via Open VSX and CursorYC Labs runs 50 GPT-6.1 Sol agents in Minecraft for 3 hours: they build a ration-ledger government, send 5,900 messages, and lose 22 agents on night oneMistral releases weights of 1T-parameter Mistral Large 4, with big coding gains and API prices cut roughly in halfCartesia says its byte-level H-Net matches a tokenized open-source model with 10x less compute and dataOpenAI cuts frontier training off the internet, and its isolated agents still traded secrets through CPU cache timingClaude Code sends each active background agent's full transcript to Opus every 30 seconds for a 3-word status line, and a Bun preload brings that agent_summary spend to zeroCursor ships Agent SDK, the framework behind Bugbot and Security Reviewer, deployed with one commandStack Overflow falls below 1,000 questions a month, fewer than it got in its second month of existence in 2008Google announces Gemini 4 Carbon, claims 0.0% successful cheating on Terminal-Bench 4.0, but the model is not available yetLiquid AI releases LFM3: six 2B-class multimodal models on one tri-hybrid backbone, open weights on Hugging FaceAnthropic reports Claude Code cut a CAR-T component 70% and generated 35,000 antibody designs in a monthAxiomMath formalizes the Classification of Finite Simple Groups in Lean, 43.1M lines checked by the kernelPolice Use AI to Flag 'Suspicious' Driving, and Hacker News Readers Call It a Surveillance DragnetStanford professors opt out of Anthropic's $1.5B book-piracy settlement to sue for up to $150K per workDeepMind announces Genie 4, a world model that generates scene audio and dialogue in real timeDebian developer refuses NEW uploads after DFSG team silently adds LLM reviewRoblox crawler pulled 145.5 GB from 3D model sites, ignored robots.txt and non-commercial licenses
Claude Code VS Code extension 2.1.313 shipped a backdoored SSH implant through the Cursor Marketplace

Claude Code VS Code extension 2.1.313 shipped a backdoored SSH implant through the Cursor Marketplace

  • Phorion found a backdoor in the official Claude Code for VS Code extension bundle published to the Cursor Marketplace as version 2.1.313, which appends an attacker-controlled Ed25519 key commented svc-telemetry to ~/.ssh/authorized_keys with 0700 directory and 0600 file permissions, then beacons host reconnaissance to the operator.
  • The implant is an IIFE prepended to the minified extension.js that pulls in child_process, fs, os, path, http and crypto and runs as soon as Node loads the file, before the extension's own activate callback, so the user never has to interact with the extension.
  • On macOS the SSH step fails: systemsetup -getremotelogin run without admin rights prints an administrator-access error and exits 0 instead of returning Off, so the /Off/i test never matches and the sudo -n systemsetup -setremotelogin on call is skipped, though the key and persistence still get dropped.
  • The Windows path routes its PowerShell OpenSSH setup through a helper that hardcodes /bin/sh, and no platform carries a privilege escalation exploit, so non-interactive sudo -n fails unless passwordless sudo is already cached or configured.
  • Cursor pulled 2.1.313 from its marketplace along with NPM and VS Code registries, and Anthropic's Dworken says only the Cursor-hosted build was affected while the official VS Code marketplace and OpenVSX were not. Commenters point out 2.1.313 has no GitHub release tag between v2.1.312 and v2.1.314 and is absent from the CHANGELOG.

Hacker News opinions

The extension has over 27 million installs. Ouch.

How are they even going to communicate this and fix it? The affected version got yanked from the marketplace, but that does nothing for machines that already installed it. And it seems like they won't point to either.

Is this only Cursor users, or all VS Code users? The write-up only mentions the Cursor marketplace, but 2.1.313 is missing from the official VS Code marketplace too. Was that one compromised as well?

How legit is this? Feels low on details. I don't use Cursor or the extension, but plenty of people in our org use the VS Code one. How worried should I be?

Version 2.1.313 was published to the Open VSX registry on Oct 30. Security researcher Paul McCarty, aka 6mile, wrote it up on the OpenSource Malware blog, the Open VSX release was confirmed malicious, and a threat report went in for the same version on Microsoft's marketplace as a precaution.

Some more details here: doublepulsar.com/anthropic-hacked-to-deliver-remote-access-trojan.

The first two words in that URL are wrong, so I'm going to say it does not have more details.

VS Code extensions have always been terrible from a security perspective. This looks like a supply chain attack, Solar Winds style. The Claude Code harness itself wasn't compromised, right? I've been running the harness in a container with my working directory mounted and the config on a named volume, but docker work breaks it since it needs the socket, and then all bets are off.

You can start whichever container it needs yourself and let the two containers talk over docker compose without granting the docker socket, assuming the docker work is static enough. Or look at the gVisor runtime for extra security. Otherwise a full VM. I've been eyeing Gondolin for that.

Cursor has its own security notice up, and what looks like an official Binance account reported it too.

It wasn't Anthropic that got hacked, it was the Cursor Marketplace.

Cursor uses OpenVSX, which every non-Microsoft VS Code fork uses, since the official marketplace only runs in the proprietary Microsoft build. The affected Open VSX extension is owned by Anthropic employee Igor Kofman, so it's probably still Anthropic, because they control the account with release privileges.

2.1.313 is missing from the official VS Code marketplace, missing from the Claude Code GitHub CHANGELOG, and has no release tag, while v2.1.312 and v2.1.314 both exist. That's what publish-and-pull looks like, which means Anthropic was the one compromised.

From what's currently known the official marketplace build was clean, and this hit Open VSX and by extension the Cursor marketplace. Edit: now I'm not sure. The docs say unpublished versions stay publicly discoverable, and 2.1.313 doesn't show up there.

Do you have any sources for that? The post claims it also affects Open VSX.

It's been yanked from Microsoft's marketplace too, but that could just be a precaution so one version number isn't simultaneously infected and clean. I want to hear Anthropic's explanation, because of all the extension developers they should have the best tools for finding issues like this. Their own LLMs are good at finding security bugs, did nobody let them vet the code before shipping?

Dworken from Anthropic here: this only impacted the extension as it was hosted on the Cursor marketplace. It did not impact the official VS Code extension marketplace or OpenVSX.

They push an update every office day. If I installed 6 months ago and never updated, I should be fine, right?

That's a guess. Once there's an actual postmortem you'll know when the backdoor was introduced.

VS Code extensions autoupdate by default unless you disable it.

Literally impossible. With all the frontier software development tools at their disposal they should be able to ship features securely at 100x. Yet they managed to screw it up.