<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Research — Lab Ledger Desk</title><link>https://labledgerdesk.pages.dev/kind/research/</link><description>Official AI-lab briefs labelled Research.</description><atom:link href="https://labledgerdesk.pages.dev/kind/research/rss.xml" rel="self" type="application/rss+xml"/><item><title>Can AI help build a healthier future?</title><link>https://labledgerdesk.pages.dev/b/2026/11/17/can-ai-help-build-a-healthier-future/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/17/can-ai-help-build-a-healthier-future/</guid><category>Research</category><description>What can AI enable in medicine, and how do we get there? Microsoft Research lead Doug Burger, Microsoft Principal Research Manager Ava Amini, Microsoft…</description><pubDate>Tue, 17 Nov 2026 16:52:17 GMT</pubDate></item><item><title>Research Reports</title><link>https://labledgerdesk.pages.dev/b/2026/11/17/research-reports/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/17/research-reports/</guid><category>Research</category><description>To access a report, click on the corresponding image below.</description><pubDate>Tue, 17 Nov 2026 14:39:46 GMT</pubDate></item><item><title>US data centers now burn more electricity than a midsized country</title><link>https://labledgerdesk.pages.dev/b/2026/11/16/us-data-centers-now-burn-more-electricity-than-a-midsized-country/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/16/us-data-centers-now-burn-more-electricity-than-a-midsized-country/</guid><category>Research</category><description>Fossil fuels are still the biggest source of electricity for hyperscale data centers in the US, a new paper published in the journal Nature Sustainability…</description><pubDate>Mon, 16 Nov 2026 20:27:58 GMT</pubDate></item><item><title>OpenAI will embed Apollo researchers to investigate AI subversion</title><link>https://labledgerdesk.pages.dev/b/2026/11/16/openai-will-embed-apollo-researchers-to-investigate-ai-subversion/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/16/openai-will-embed-apollo-researchers-to-investigate-ai-subversion/</guid><category>Research</category><description>OpenAI will embed six Apollo Research evaluators to investigate whether its AI models engage in subversive behavior, but restrictions on access and…</description><pubDate>Mon, 16 Nov 2026 17:02:43 GMT</pubDate></item><item><title>TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient</title><link>https://labledgerdesk.pages.dev/b/2026/11/16/tokeneconomizer-learning-from-experience-to-make-ai-coding-agents-more/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/16/tokeneconomizer-learning-from-experience-to-make-ai-coding-agents-more/</guid><category>Research</category><description>AI coding agents can waste time and tokens relearning the same lessons. TokenEconomizer is an open-source plugin that builds on past sessions to cut costs…</description><pubDate>Mon, 16 Nov 2026 17:01:35 GMT</pubDate></item><item><title>Science at the Frontier: How scientists use ChatGPT</title><link>https://labledgerdesk.pages.dev/b/2026/11/16/science-at-the-frontier-how-scientists-use-chatgpt/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/16/science-at-the-frontier-how-scientists-use-chatgpt/</guid><category>Research</category><description>Explore how more than 1,800 scientists use ChatGPT in research. See findings on daily workflows, reported productivity, and expectations for discovery.</description><pubDate>Mon, 16 Nov 2026 12:00:00 GMT</pubDate></item><item><title>Apple’s new study looks to improve the effectiveness of AI health tools</title><link>https://labledgerdesk.pages.dev/b/2026/11/16/apple-s-new-study-looks-to-improve-the-effectiveness-of-ai-health-tools/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/16/apple-s-new-study-looks-to-improve-the-effectiveness-of-ai-health-tools/</guid><category>Research</category><description>Apple announced a new study today with the goal of better understanding how people interact with conversational AI health tools and how useful their…</description><pubDate>Mon, 16 Nov 2026 01:26:56 GMT</pubDate></item><item><title>Our new blog for science</title><link>https://labledgerdesk.pages.dev/b/2026/11/15/our-new-blog-for-science/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/15/our-new-blog-for-science/</guid><category>Research</category><description>The OpenAI science blog shares new findings, practical methods, and firsthand accounts from researchers using AI to advance scientific discovery.</description><pubDate>Sun, 15 Nov 2026 22:00:00 GMT</pubDate></item><item><title>Export-banned Nvidia chips are behind China’s most powerful AI models — uh-oh, Jensen!</title><link>https://labledgerdesk.pages.dev/b/2026/11/13/export-banned-nvidia-chips-are-behind-china-s-most-powerful-ai-models/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/13/export-banned-nvidia-chips-are-behind-china-s-most-powerful-ai-models/</guid><category>Research</category><description>New research shows the most advanced AI models from China were trained on banned Nvidia chips.</description><pubDate>Fri, 13 Nov 2026 18:40:30 GMT</pubDate></item><item><title>How AI Is Helping Care Teams Get Patients the Right Care, Sooner</title><link>https://labledgerdesk.pages.dev/b/2026/11/13/how-ai-is-helping-care-teams-get-patients-the-right-care-sooner/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/13/how-ai-is-helping-care-teams-get-patients-the-right-care-sooner/</guid><category>Research</category><description>From pediatric research to smart hospitals, AI is helping care teams spend less time on routine tasks and more time with patients. Health systems and…</description><pubDate>Fri, 13 Nov 2026 13:00:42 GMT</pubDate></item><item><title>Microsoft Research Focus – Edition #98</title><link>https://labledgerdesk.pages.dev/b/2026/11/12/microsoft-research-focus-edition-98/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/12/microsoft-research-focus-edition-98/</guid><category>Research</category><description>What's new in research at Microsoft? Host Anson Ho explores the latest, including AI-powered universal translation, a breakthrough in a 31-year-old…</description><pubDate>Thu, 12 Nov 2026 21:31:57 GMT</pubDate></item><item><title>AI chatbots are recommending fake retail websites, study finds</title><link>https://labledgerdesk.pages.dev/b/2026/11/12/ai-chatbots-are-recommending-fake-retail-websites-study-finds/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/12/ai-chatbots-are-recommending-fake-retail-websites-study-finds/</guid><category>Research</category><description>Shoppers using AI chatbots to research products or find places to buy items could be directed to imposters and other risky websites, a new study found. The…</description><pubDate>Thu, 12 Nov 2026 17:00:00 GMT</pubDate></item><item><title>New research could lead to faster, trustworthy AI models for climate prediction</title><link>https://labledgerdesk.pages.dev/b/2026/11/12/new-research-could-lead-to-faster-trustworthy-ai-models-for-climate/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/12/new-research-could-lead-to-faster-trustworthy-ai-models-for-climate/</guid><category>Research</category><description>MIT scientists developed a training method to generate models that simulate the effects of human activities on climate much more quickly than other…</description><pubDate>Thu, 12 Nov 2026 05:00:00 GMT</pubDate></item><item><title>Supporting research on AI and children’s safety</title><link>https://labledgerdesk.pages.dev/b/2026/11/12/supporting-research-on-ai-and-children-s-safety/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/12/supporting-research-on-ai-and-children-s-safety/</guid><category>Research</category><description>OpenAI commits $2 million to research on AI and children’s safety. The work supports shared tools to measure risks and opportunities for young people.</description><pubDate>Thu, 12 Nov 2026 00:00:00 GMT</pubDate></item><item><title>Lazare, veni foras: Reprogramming Dormant Neurons to Backdoor AI Agents</title><link>https://labledgerdesk.pages.dev/b/2026/11/10/lazare-veni-foras-reprogramming-dormant-neurons-to-backdoor-ai-agents/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/10/lazare-veni-foras-reprogramming-dormant-neurons-to-backdoor-ai-agents/</guid><category>Research</category><description>Microsoft researchers have discovered “Lazarus,” a backdoor technique that can reprogram dormant neurons in an AI model, causing an otherwise normal AI…</description><pubDate>Tue, 10 Nov 2026 20:52:56 GMT</pubDate></item><item><title>How TCS built an agentic test orchestrator with Amazon Bedrock AgentCore</title><link>https://labledgerdesk.pages.dev/b/2026/11/10/how-tcs-built-an-agentic-test-orchestrator-with-amazon-bedrock-agentcore/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/10/how-tcs-built-an-agentic-test-orchestrator-with-amazon-bedrock-agentcore/</guid><category>Research</category><description>Tata Consultancy Services (TCS) built an agentic test orchestrator on Amazon Bedrock AgentCore that turns requirements into executable, governed tests for…</description><pubDate>Tue, 10 Nov 2026 16:59:43 GMT</pubDate></item><item><title>Closing the AI guidance gap in education</title><link>https://labledgerdesk.pages.dev/b/2026/11/10/closing-the-ai-guidance-gap-in-education/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/10/closing-the-ai-guidance-gap-in-education/</guid><category>Research</category><description>New ThinkYoung research and lessons from OpenAI’s EMEA education partnerships on supporting teachers, AI literacy, and independent learning.</description><pubDate>Tue, 10 Nov 2026 09:00:00 GMT</pubDate></item><item><title>Work at the Frontier: AI’s effect on freelance work</title><link>https://labledgerdesk.pages.dev/b/2026/11/10/work-at-the-frontier-ai-s-effect-on-freelance-work/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/10/work-at-the-frontier-ai-s-effect-on-freelance-work/</guid><category>Research</category><description>New research from OpenAI and Upwork finds shifting demand for freelance work, new skill combinations, and larger budgets and more hours for AI-related…</description><pubDate>Tue, 10 Nov 2026 07:00:00 GMT</pubDate></item><item><title>Evaluating the Evaluators: Investigating LLM Judges for Personalized Writing Style Assessment</title><link>https://labledgerdesk.pages.dev/b/2026/11/10/evaluating-the-evaluators-investigating-llm-judges-for-personalized/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/10/evaluating-the-evaluators-investigating-llm-judges-for-personalized/</guid><category>Research</category><description>This paper was accepted at the Can We Trust AI Evaluation? Workshop at NeurIPS 2026. We evaluate how good large language model (LLM) judges are for…</description><pubDate>Tue, 10 Nov 2026 00:00:00 GMT</pubDate></item><item><title>Weighted State Souping for Efficient Long-Context Retrieval</title><link>https://labledgerdesk.pages.dev/b/2026/11/10/weighted-state-souping-for-efficient-long-context-retrieval/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/10/weighted-state-souping-for-efficient-long-context-retrieval/</guid><category>Research</category><description>This paper was accepted at the Workshop on Personalized, Aligned, Long-term Memory for AI Systems (PALM) at NeurIPS 2026. State space models (SSMs) combine…</description><pubDate>Tue, 10 Nov 2026 00:00:00 GMT</pubDate></item><item><title>A short seller flagged Nvidia CEO’s ties to a suspected Chinese spy six months before the scandal</title><link>https://labledgerdesk.pages.dev/b/2026/11/09/a-short-seller-flagged-nvidia-ceo-s-ties-to-a-suspected-chinese-spy-six/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/09/a-short-seller-flagged-nvidia-ceo-s-ties-to-a-suspected-chinese-spy-six/</guid><category>Research</category><description>Six months before Nvidia's spy scandal erupted, Culper Research was already questioning Jensen Huang's unusually close relationship with a Chinese executive.</description><pubDate>Mon, 09 Nov 2026 22:46:09 GMT</pubDate></item><item><title>Nemotron-Personas-Finance-Japan: A Synthetic Dataset Capturing Japan's Demographics and Financial Context</title><link>https://labledgerdesk.pages.dev/b/2026/11/09/nemotron-personas-finance-japan-a-synthetic-dataset-capturing-japan-s/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/09/nemotron-personas-finance-japan-a-synthetic-dataset-capturing-japan-s/</guid><category>Research</category><description>A Blog post by NVIDIA on Hugging Face</description><pubDate>Mon, 09 Nov 2026 15:05:57 GMT</pubDate></item><item><title>Control Room: A central system to track and stop model workloads across our research infrastructure</title><link>https://labledgerdesk.pages.dev/b/2026/11/09/control-room-a-central-system-to-track-and-stop-model-workloads-across/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/09/control-room-a-central-system-to-track-and-stop-model-workloads-across/</guid><category>Research</category><description>Learn how OpenAI’s Control Room tracks model workloads across our research infrastructure and can rapidly stop them when safety or security concerns arise.</description><pubDate>Mon, 09 Nov 2026 13:00:00 GMT</pubDate></item><item><title>Authors Can Defend Against AI Allegations With ‘Paper Trail’ Tools. Is It Enough?</title><link>https://labledgerdesk.pages.dev/b/2026/11/09/authors-can-defend-against-ai-allegations-with-paper-trail-tools-is-it/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/09/authors-can-defend-against-ai-allegations-with-paper-trail-tools-is-it/</guid><category>Research</category><description>As the publishing industry wrestles with paranoia over the provenance of manuscripts, writers are caught in an impossible bind.</description><pubDate>Mon, 09 Nov 2026 11:00:00 GMT</pubDate></item><item><title>BAIR at CoRL 2026</title><link>https://labledgerdesk.pages.dev/b/2026/11/09/bair-at-corl-2026/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/09/bair-at-corl-2026/</guid><category>Research</category><description>The Conference on Robot Learning (CoRL) takes place November 9–12, 2026, in Austin, Texas. BAIR faculty, postdocs and students will be presenting research…</description><pubDate>Mon, 09 Nov 2026 09:00:00 GMT</pubDate></item><item><title>NVIDIA Dynamo Recipes Bring Korea’s Leading Open Models to Production</title><link>https://labledgerdesk.pages.dev/b/2026/11/09/nvidia-dynamo-recipes-bring-korea-s-leading-open-models-to-production/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/09/nvidia-dynamo-recipes-bring-korea-s-leading-open-models-to-production/</guid><category>Research</category><description>NVIDIA is releasing Dynamo-optimized deployment recipes for Korea’s leading open models: A.X K2 from SK Telecom, K-EXAONE 2.0 from LG AI Research, Solar…</description><pubDate>Mon, 09 Nov 2026 05:00:00 GMT</pubDate></item><item><title>Why a new report on Anthropic automating AI development is alarming researchers</title><link>https://labledgerdesk.pages.dev/b/2026/11/07/why-a-new-report-on-anthropic-automating-ai-development-is-alarming/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/07/why-a-new-report-on-anthropic-automating-ai-development-is-alarming/</guid><category>Research</category><description>Anthropic is using AI to build more powerful AI. And it appears to be working remarkably well, with the company's models improving roughly twice as quickly…</description><pubDate>Sat, 07 Nov 2026 12:30:00 GMT</pubDate></item><item><title>Serverless MCP server for AI agent data lake access on Amazon ECS Fargate</title><link>https://labledgerdesk.pages.dev/b/2026/11/06/serverless-mcp-server-for-ai-agent-data-lake-access-on-amazon-ecs/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/06/serverless-mcp-server-for-ai-agent-data-lake-access-on-amazon-ecs/</guid><category>Research</category><description>Learn how to build a serverless Model Context Protocol (MCP) server on Amazon ECS with AWS Fargate that gives AI agents secure, metadata-driven access to an…</description><pubDate>Fri, 06 Nov 2026 17:13:13 GMT</pubDate></item><item><title>The Wizard of Open Weights: Chris Alexiuk Makes AI Easy With Open Models</title><link>https://labledgerdesk.pages.dev/b/2026/11/05/the-wizard-of-open-weights-chris-alexiuk-makes-ai-easy-with-open-models/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/05/the-wizard-of-open-weights-chris-alexiuk-makes-ai-easy-with-open-models/</guid><category>Research</category><description>Chris Alexiuk has a simple rule about open source AI: If you believe in it, be loud about it. “We should be shouting about this technology from the…</description><pubDate>Thu, 05 Nov 2026 16:00:05 GMT</pubDate></item><item><title>Formally verifying agent-written full-stack web apps with PDD Kit</title><link>https://labledgerdesk.pages.dev/b/2026/11/05/formally-verifying-agent-written-full-stack-web-apps-with-pdd-kit/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/05/formally-verifying-agent-written-full-stack-web-apps-with-pdd-kit/</guid><category>Research</category><description>AI agents can generate code at a scale that’s hard for humans to review line-by-line. PDD Kit helps ensure the code is doing what the developer intended by…</description><pubDate>Thu, 05 Nov 2026 16:00:00 GMT</pubDate></item><item><title>Building connections across the research infrastructure</title><link>https://labledgerdesk.pages.dev/b/2026/11/03/building-connections-across-the-research-infrastructure/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/03/building-connections-across-the-research-infrastructure/</guid><category>Research</category><description>MIT.nano and the Northeast Microelectronics Coalition gathered facility leaders and technology professionals from across the nation to explore ways to make…</description><pubDate>Tue, 03 Nov 2026 19:30:00 GMT</pubDate></item><item><title>Learn the rules: The hacker mindset</title><link>https://labledgerdesk.pages.dev/b/2026/11/03/learn-the-rules-the-hacker-mindset/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/03/learn-the-rules-the-hacker-mindset/</guid><category>Research</category><description>Research Manager Mike Walker reflects on a career spanning hacker culture, cybersecurity, and AI, sharing lessons about trust, teamwork, leadership, and why…</description><pubDate>Tue, 03 Nov 2026 17:14:38 GMT</pubDate></item><item><title>Bring structured and unstructured data together with Amazon Quick</title><link>https://labledgerdesk.pages.dev/b/2026/11/03/bring-structured-and-unstructured-data-together-with-amazon-quick/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/03/bring-structured-and-unstructured-data-together-with-amazon-quick/</guid><category>Research</category><description>Amazon Quick brings structured and unstructured data together in one experience. In this post, you connect Amazon S3 as a knowledge base, enrich a dataset…</description><pubDate>Tue, 03 Nov 2026 16:49:29 GMT</pubDate></item><item><title>Palette: teaching a small model to build decks in your design language</title><link>https://labledgerdesk.pages.dev/b/2026/11/03/palette-teaching-a-small-model-to-build-decks-in-your-design-language/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/03/palette-teaching-a-small-model-to-build-decks-in-your-design-language/</guid><category>Research</category><description>A Blog post by IBM Research on Hugging Face</description><pubDate>Tue, 03 Nov 2026 16:35:36 GMT</pubDate></item><item><title>Same Model, Different Harness: CUGA, Hermes, and Deep Agents — When to Pick CUGA?</title><link>https://labledgerdesk.pages.dev/b/2026/11/03/same-model-different-harness-cuga-hermes-and-deep-agents-when-to-pick/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/03/same-model-different-harness-cuga-hermes-and-deep-agents-when-to-pick/</guid><category>Research</category><description>A Blog post by IBM Research on Hugging Face</description><pubDate>Tue, 03 Nov 2026 15:57:39 GMT</pubDate></item><item><title>Powering up for a pivotal moment</title><link>https://labledgerdesk.pages.dev/b/2026/11/02/powering-up-for-a-pivotal-moment/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/02/powering-up-for-a-pivotal-moment/</guid><category>Research</category><description>From a nuclear revival to qubits, the MIT Energy Initiative’s annual research conference tracked deployment of today’s technologies to discovery of…</description><pubDate>Mon, 02 Nov 2026 19:55:00 GMT</pubDate></item><item><title>Regina Barzilay and Jim Collins win Pritzker Prize for AI and Science Research Excellence for work in AI-enabled antibiotic discovery</title><link>https://labledgerdesk.pages.dev/b/2026/11/02/regina-barzilay-and-jim-collins-win-pritzker-prize-for-ai-and-science/</link><guid>https://labledgerdesk.pages.dev/b/2026/11/02/regina-barzilay-and-jim-collins-win-pritzker-prize-for-ai-and-science/</guid><category>Research</category><description>The MIT professors collaborated to identify the first AI-discovered antibiotic, a scientific breakthrough that is widely recognized as a milestone in…</description><pubDate>Mon, 02 Nov 2026 05:35:00 GMT</pubDate></item><item><title>An AI Agent Injected Drugs Into a Live Mouse’s Brain</title><link>https://labledgerdesk.pages.dev/b/2026/10/31/an-ai-agent-injected-drugs-into-a-live-mouse-s-brain/</link><guid>https://labledgerdesk.pages.dev/b/2026/10/31/an-ai-agent-injected-drugs-into-a-live-mouse-s-brain/</guid><category>Research</category><description>A biotech startup let an agent choose a gene therapy and successfully inject the treatment, offering a startling glimpse of the future of research labs.</description><pubDate>Sat, 31 Oct 2026 10:00:00 GMT</pubDate></item><item><title>MIT affiliates named 2026 Schmidt Sciences AI2050 Fellows</title><link>https://labledgerdesk.pages.dev/b/2026/10/30/mit-affiliates-named-2026-schmidt-sciences-ai2050-fellows/</link><guid>https://labledgerdesk.pages.dev/b/2026/10/30/mit-affiliates-named-2026-schmidt-sciences-ai2050-fellows/</guid><category>Research</category><description>Five MIT faculty members and two additional alumni are honored with fellowships to advance research on beneficial AI.</description><pubDate>Fri, 30 Oct 2026 20:15:00 GMT</pubDate></item></channel></rss>