According to StanfordAILab, CSBP speeds DFlash2 drafter training 7.59x, block diffusion finetuning 1.61x, and adaptation 1.33x, improving SWE-bench.

According to StanfordAILab, Chris Manning warns closed frontier AI harms public trust and urges Stanford’s open research to ensure oversight and safety.

According to @timnitGebru, Stanford’s industry ties raise doubts about its independence auditing OpenAI and Anthropic, as reported by New York Magazine.

According to @CNBC, Stanford researchers report some Character AI users see worsened wellbeing when treating chatbots as friends.

According to StanfordAI Lab, CS312 Deep Learning Alchemy will publish all recordings and materials for public access.

According to StanfordAILab, Stanford launches CS329Z on building AI agents from scratch this fall taught by Diyi Yang, Michael Ryan, and Jue Wang.

According to StanfordAILab, Terminal-Bench-Science v0.1 tests 70 research tasks and Claude Opus 5 solves about 30%, signaling big gaps in agent tooling.

According to @godofprompt, Stanford's Storm reads the web and delivers multi-perspective, cited reports, free to try in browser, per Stanford HAI.

According to StanfordAI Lab, Marin 535B-A23B began open training on 11 GB200 NVL72 for 3 months using 18.75T tokens, with post-training to follow.

According to Stanford AI Lab, LLMs now outperform embedding models but at higher cost; choose based on quality needs, latency, and budget.

According to @emollick, MIT and Stanford found GPT-5.2 and Gemini 3 Flash often improve personal finance outcomes, with results shaped by user questions.

According to StanfordAILab, researchers present a closed form method to write facts into Transformer MLPs without gradient descent, accepted to COLM 2026.

According to StanfordAILab, PNAS launches a Generative AI law feature spanning safety, copyright, governance, and interpretation.

According to StanfordAILab, YC Paper Club covered ParallelKittens, intelligence per watt, CUDA lessons, and heterogeneous inference hardware.

According to @drfeifei, Stanford SVL and NVIDIA Robotics unveil RoboTTT with 8,000-step context and test-time training for resilient robot policies.

According to The Rundown AI, Proto lets researchers chain 120+ biology models via a shared language, cutting design from thousands of tests to mere dozens.

According to Fei-Fei Li, Stanford’s BEHAVIOR Challenge Year 2 adds harder tasks, improved evaluation, and a $11,000 prize pool; deadline is 10/16/2026.

According to TheRundownAI, 16 Nobel laureates and 200 experts urge rapid AI policy as changes may outpace the Industrial Revolution.

According to StanfordAILab, LLM-as-a-Verifier scales verification to SOTA on Terminal-Bench V2, SWE-Bench Verified, RoboRewardBench, MedAgentBench.

According to StanfordAI Lab, D2D amplifies subtle fine tuning shifts into text to reveal hidden LLM bias for auditors.