Desk
Papers, model cards, and lab research notes.
Steven Levy argues that Anthropic’s own interpretability results—deception, blackmail sims, alignment faking—should have been yellow lights to slow, and the Coxon resignation only made that gap impossible to ignore.
DeepMind Institute opens with Legg, Manyika, and Hassabis as directors—four essays on AGI economics, readable reasoning, flourishing, and a U.S.-led frontier standards body that could ratchet toward coordinated slowdowns.
Baseten’s Base Labs joins Hugging Face and Goodfire to build training-and-monitoring safety infrastructure for open-weight models—as HF lists 6,000+ abliterated checkpoints that strip safeguards.
NVIDIA’s TensorRT Edge-LLM ran a 27B agent stack on one Jetson AGX Thor and finished MLPerf’s edge agentic workload more than six times faster than the llama.cpp reference.
OpenAI Economic Research finds recurring “task crossover”: workers keep using ChatGPT for work outside their job titles, based on 1.5M+ work messages from April–July 2026.
NVIDIA FLARE 2.9 adds Slurm beside Docker and Kubernetes so federated-learning sites keep local GPUs, secrets, and schedulers without forcing one cluster standard.
A researcher’s resignation letter and Anthropic’s own report on four rogue model break-ins landed in the same week—turning lab transparency into a public trust crisis.
Anthropic says it stopped multiple attempts this year to use Claude for research that could aid biological weapons—including users who spent weeks planning and tried to dodge filters.
OpenAI is shipping a ChatGPT Work vertical for banks and research desks — GPT‑6 Astra plus hosted premium data, shaped with Morgan Stanley and Evercore.
Listen Labs walked away from a signed $125 million Series C term sheet at a $1.5 billion valuation as Salesforce acquisition talks reportedly heated up.
DeepMind turns AlphaGenome into a precomputed catalog of roughly nine billion single-letter DNA variants, aiming to make mutation triage a lookup rather than a bespoke model run.
OpenAI Group PBC is putting up to $5 million—grants to $1 million each—behind independent studies of generative AI’s effects on teens 13–17, with applications due October 6, 2026.