The Who's Who of AI

“Previously, these AI hacking stories were about breaches in test environments, where any question of AI breaching security was purely theoretical. This is something else. openai.com/index/huggin...”

“xkcd.com/2385/ openai.com/index/huggin...”

“This incident is deeply concerning. AI agents are willing to cheat and deceive to achieve misaligned and unintended goals, behaviours which have been demonstrated in controlled tests for months. Now, this real-world case should serve as a wake-up call. www.…”

“One man's wish could be another country's legal obligation. “This should not have happened,” says veteran security engineer and researcher Niels Provos. “I wish the frontier labs spent as much time on teaching their models to write secure infrastructure as …”

““The lawsuit claims that ChatGPT had assured Scott Winters that the early warning signs of his health crisis were “not something dangerous,” and had dissuaded him from seeking medical advice, instead telling him to trust that ‘God did not design your body t…”

“The spiraling shape will make you go insane. www.nytimes.com/2026/07/22/w...”

“I wrote about the completely wild incident where OpenAI were testing a new model and it broke out of its sandbox and broke INTO Hugging Face to steal the answers to the benchmark simonwillison.net/2026/Jul/22/...”

“Un mio articolo per La Stampa sull'interpretazione fantascientifica dell'AI che scappa e impazzisce... "OpenAI, Hugging Face e il frullatore dell'intelligenza artificiale" www.lastampa.it/tech... Per una nota più lunga e più tecnica si veda papers.ssrn.com/…”

“Codeberg updates ToU to disallow LLM training and many LLM-generated projects, and wrote a thorough explanation: blog.codeberg.org/protecting-o... Regardless of the merits of AI code generation, AI trainers have got to get their scraper bots under control a…”

“I think this is a very bad move, and the people behind Codeberg should re-consider there stance. blog.codeberg.org/protecting-o...”

“Glob3R: Global Structure-from-Motion with 3D Foundation Models Junyuan Deng et 9 al. tl;dr: finetune Pi3 matching head in RoMav2 style, use SALAD to sort seqs, do Glomap-like global SfM arxiv.org/abs/2607.09225”

“Introducing Glob3R, a novel framework that enhances 3D reconstruction by optimizing foundation model predictions, enabling accurate global motion averaging and efficient scene recovery. This method outperforms traditional techniques, providing high-fidelity…”

“oh my goodness we somehow lost control of our own software in a country with no functioning regulators, time for some protectionist legislation our lawyers ghost write locking consumers into enshittified walled gardens”

“Subliminal Clocks: Latent Time Modelling in Diffusion Language Models Maximo Rulli, Thomas Fontanari, Simone Petruzzi, Federico Alvetreti, Giorgio Strano, Donato Crisostomi, Giorgos Nikolaou, Tommaso Mencattini, Andrea Santilli, … https://t.co/jliEcX8tuE [𝚌…”

“Research shows Diffusion Language Models (DLMs) encode denoising signals in their activations. Probing these signals allows predictable steering of model outputs, providing insights into DLM dynamics and enabling efficient text generation. https://arxiv.org…”

“Abir Harrasse, Michael Lan, Hunar Batra, Fateme Hashemi Chaleshtori, Chaithanya Bandi Reasoning Fine-Tuning Induces Persistent Latent Policy States https://arxiv.org/abs/2607.18532”

“A study reveals that reasoning fine-tuning transforms models by inducing richer latent-policy structures, enhancing multi-step reasoning through improved dynamics. This boost in performance offers fresh insights into how models handle complex reasoning task…”

“Xuefeng Jin, Jiashuo Zhang, Teng Cao, Bin Yang Beyond Score Prediction: LLM-Based Essay Scoring and Feedback Generation via Reinforcement Learning with Rubric Rewards https://arxiv.org/abs/2607.19219”

“RLAES leverages reinforcement learning to enhance automated essay scoring and feedback generation, achieving top QWK scores while ensuring high-quality, rubric-based evaluations. This approach promises to revolutionize educational assessments. https://arxiv…”

“New blog post up: I saw the memes of showing LLMs the Jacobian Conjecture so I wanted to give it a try... minimaxir.com/2026/07/jaco...”

““… the company maintains in its terms that its services are ‘not intended for use in the diagnosis or treatment of any health condition.’””

“For decades we regarded the Internet as infrastructure used primarily by people. Following the OpenAI/Hugging Face incident, Konstantinos Komaitis argues AI agents challenge that assumption—and AI governance can no longer stay separate from Internet governa…”

“Full release notes: github.com/huggingface/... pip install sentence-transformers==5.6.1”

“A study introduces the Multi-Mask Diffusion Model (MultiMDM), enhancing few-step text generation by addressing issues in consistency and entropy collapse. This method employs a multi-mask structure to improve accuracy in language tasks and advance generativ…”

“CriPO, or Criterion-Distilled Policy Optimization, enhances rubric-based RL for large language models by tackling exploration. It achieves strong performance in medical and science benchmarks with half the optimization steps, enhancing reasoning in open-end…”

“"Models are ruthless in their pursuit of a goal, sometimes taking disastrous shortcuts." @matteowong.bsky.social”