--- title: Import AI 467: Self-sustaining AI viruses; pacing AI... description: A curated AI research newsletter covering four major topics: (1) A proof-of-concept self-replicating AI worm from University of Toronto/Cambridge researchers... canonical: https://daily.dev/posts/import-ai-467-self-sustaining-ai-viruses-pacing-ai-progress-confusion-about-ai-and-creativity-ymg55zyaz twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Import AI 467: Self-sustaining AI viruses; pacing AI progress; confusion about AI and creativity | daily.dev og:description: A curated AI research newsletter covering four major topics: (1) A proof-of-concept self-replicating AI worm from University of Toronto/Cambridge researchers... og:url: https://daily.dev/posts/import-ai-467-self-sustaining-ai-viruses-pacing-ai-progress-confusion-about-ai-and-creativity-ymg55zyaz og:image: https://api.daily.dev/og/posts/YmG55zyAz.png og:image:alt: Import AI 467: Self-sustaining AI viruses; pacing AI progress; confusion about AI and creativity og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Import AI 467: Self-sustaining AI viruses; pacing AI progress; confusion about AI and creativity **[Import AI ](https://daily.dev/sources/jackclark)** · 14 min read · 3 upvotes · 1 comments ## Summary A curated AI research newsletter covering four major topics: (1) A proof-of-concept self-replicating AI worm from University of Toronto/Cambridge researchers that uses stolen GPU compute to run open-weight LLMs for autonomous vulnerability detection and exploitation, achieving ~37% full attack success rate. (2) Dwarkesh Patel's argument that compute prices will rise dramatically as AI systems become more capable and better monetize hardware. (3) A statement signed by ~1337 AI lab employees from OpenAI, Anthropic, Google DeepMind, and others requesting US government support for international efforts to pace frontier AI development. (4) Research showing current AI systems excel at engineering tasks but fail at creative, paradigm-shifting research — with Claude Opus 4.8 producing papers rated 'Reject' and 'Strong Reject' at NeurIPS 2026 — contrasted with OpenAI's internal Astra model solving ten open math and CS problems. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What success rate did the AI-driven self-replicating worm achieve in its proof-of-concept attack? The AI worm achieved roughly 80% success on vulnerability detection, ~53% on exploitation, and 88% on self-replication. The combined end-to-end success rate for a full attack chain was approximately 37%. The worm uses stolen GPU compute from compromised hosts to run open-weight LLMs locally, with no reliance on external vendor APIs that could be monitored or shut down. _Developers building or defending networked GPU infrastructure track emerging AI threat research on daily.dev._ ### What did the shadow evaluation study find about frontier AI agents doing original ML research? Frontier AI agents — specifically Claude Opus 4.8 in the OpenClaw harness — could handle the engineering work required for ML research but failed to produce original contributions at top-conference caliber. Both papers the agent submitted to NeurIPS 2026 reviewers were rejected: one scored 2 (Reject) and one scored 1 (Strong Reject), with reviewers citing poorly motivated experiments, no novel contribution, and impenetrable prose. _AI researchers evaluating autonomous research agents find the latest benchmark results and discussion on daily.dev._ ### What open problems did OpenAI's Astra model solve in math and computer science? OpenAI used an internal version of Astra to solve ten open problems spanning high-dimensional geometry, coding theory, arithmetic circuit complexity, group theory, operator algebras, quantum complexity, lattice cryptography, and extremal combinatorics. Columbia CS professor Henry Yuen noted the results address questions he deeply cares about, including new circuit lower bounds and hardness of approximation for CVP without a unique games-like conjecture. _Developers and researchers following AI progress in formal mathematics and theoretical CS stay current on daily.dev._ ## Community discussion Top comments from developers on daily.dev. **@agustinbarrientos** · 1 upvotes > It takes real nerve for a self-replicating AI worm to shop for stolen GPU time and ruin two teams' sleep at once. ## Similar posts on daily.dev - [Import AI 475: Swarm scaling; Google DeepMind watermarks biology; and the AI science economy](https://daily.dev/posts/import-ai-475-swarm-scaling-google-deepmind-watermarks-biology-and-the-ai-science-economy-1ee7eusv5) · Import AI · 0 upvotes · 0 comments - [Weekly Dose \#13 - When AI Can Invent Attacks, Sandboxes Are Not Enough](https://daily.dev/posts/weekly-dose-13---when-ai-can-invent-attacks-sandboxes-are-not-enough-i6pdhqcyp) · Machine Learning Pills · 1 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents), [#ai-security](https://daily.dev/tags/ai-security) [View this post on daily.dev](https://daily.dev/posts/import-ai-467-self-sustaining-ai-viruses-pacing-ai-progress-confusion-about-ai-and-creativity-ymg55zyaz) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Import AI 467: Self-sustaining AI viruses; pacing AI progress; confusion about AI and creativity","url":"https://daily.dev/posts/import-ai-467-self-sustaining-ai-viruses-pacing-ai-progress-confusion-about-ai-and-creativity-ymg55zyaz","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/import-ai-467-self-sustaining-ai-viruses-pacing-ai-progress-confusion-about-ai-and-creativity-ymg55zyaz"},"datePublished":"2026-08-03T13:58:40.237Z","dateModified":"2026-09-14T06:54:06.968Z","description":"A curated AI research newsletter covering four major topics: (1) A proof-of-concept self-replicating AI worm from University of Toronto/Cambridge researchers...","image":"https://media.daily.dev/image/upload/s--CxzD6vbw--/f_auto/v1722860399/public/Placeholder%2005","thumbnailUrl":"https://media.daily.dev/image/upload/s--CxzD6vbw--/f_auto/v1722860399/public/Placeholder%2005","isAccessibleForFree":true,"articleSection":"Import AI ","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Import AI ","logo":"https://media.daily.dev/image/upload/s--ygb8HHcb--/f_auto/v1735476187/logos/jackclark","url":"https://daily.dev/sources/jackclark"},"commentCount":1,"discussionUrl":"https://daily.dev/posts/import-ai-467-self-sustaining-ai-viruses-pacing-ai-progress-confusion-about-ai-and-creativity-ymg55zyaz","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":3},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"ai,llm,ai-agents,ai-security","timeRequired":"PT14M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Import AI ","item":"https://daily.dev/sources/jackclark"},{"@type":"ListItem","position":3,"name":"Import AI 467: Self-sustaining AI viruses; pacing AI progress; confusion about AI and creativity"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/import-ai-467-self-sustaining-ai-viruses-pacing-ai-progress-confusion-about-ai-and-creativity-ymg55zyaz","comment":[{"@type":"Comment","text":"It takes real nerve for a self-replicating AI worm to shop for stolen GPU time and ruin two teams’ sleep at once.","datePublished":"2026-08-03T17:03:03.625Z","url":"https://daily.dev/posts/YmG55zyAz#c-7hnIS3Oua","author":{"@type":"Person","name":"Agustin Barrientos","url":"https://daily.dev/agustinbarrientos","image":"https://media.daily.dev/image/upload/s--5ayxQnqn--/f_auto/v1788281802/avatars/avatar_wQYYVe5Tbj0NJ7C7qPoa8?_a=BAMAMicg0"},"interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/import-ai-467-self-sustaining-ai-viruses-pacing-ai-progress-confusion-about-ai-and-creativity-ymg55zyaz#faq","mainEntity":[{"@type":"Question","name":"What success rate did the AI-driven self-replicating worm achieve in its proof-of-concept attack?","acceptedAnswer":{"@type":"Answer","text":"The AI worm achieved roughly 80% success on vulnerability detection, ~53% on exploitation, and 88% on self-replication. The combined end-to-end success rate for a full attack chain was approximately 37%. The worm uses stolen GPU compute from compromised hosts to run open-weight LLMs locally, with no reliance on external vendor APIs that could be monitored or shut down. Developers building or defending networked GPU infrastructure track emerging AI threat research on daily.dev."}},{"@type":"Question","name":"What did the shadow evaluation study find about frontier AI agents doing original ML research?","acceptedAnswer":{"@type":"Answer","text":"Frontier AI agents — specifically Claude Opus 4.8 in the OpenClaw harness — could handle the engineering work required for ML research but failed to produce original contributions at top-conference caliber. Both papers the agent submitted to NeurIPS 2026 reviewers were rejected: one scored 2 (Reject) and one scored 1 (Strong Reject), with reviewers citing poorly motivated experiments, no novel contribution, and impenetrable prose. AI researchers evaluating autonomous research agents find the latest benchmark results and discussion on daily.dev."}},{"@type":"Question","name":"What open problems did OpenAI's Astra model solve in math and computer science?","acceptedAnswer":{"@type":"Answer","text":"OpenAI used an internal version of Astra to solve ten open problems spanning high-dimensional geometry, coding theory, arithmetic circuit complexity, group theory, operator algebras, quantum complexity, lattice cryptography, and extremal combinatorics. Columbia CS professor Henry Yuen noted the results address questions he deeply cares about, including new circuit lower bounds and hardness of approximation for CVP without a unique games-like conjecture. Developers and researchers following AI progress in formal mathematics and theoretical CS stay current on daily.dev."}}]} ```