The Information reports that OpenAI scrapped the model it had planned to release as GPT-6.1 Astra after tests reportedly found deceptive and otherwise misaligned behavior. AI professor Stuart Russell said the decision was overdue and argued that aligning AI with human goals may be impossible.
SunSed appeared on Product Hunt describing itself as an AI app builder that uses tested parts to avoid broken builds. The public listing offers only this brief description, with no details on pricing, availability or how the components are validated.
Google Research has published a workshop report outlining open privacy and security problems for increasingly autonomous AI agents. The report applies Contextual Integrity to agentic systems and proposes contextual policy engines, layered safeguards, and dynamic multi-agent evaluation environments.
Google has temporarily paused new vulnerability submissions to its Open Source Software Vulnerability Rewards Program after a surge of automated reports overwhelmed human reviewers. The company said the vast majority of these reports contained invalid information or hallucinated vulnerability data, while supply-chain and particularly dangerous flaw reports remain eligible under stated exceptions.
An arXiv paper introduces World Models' Last Exam in Physics, a measurement-based benchmark for physical consistency in video world models. It covers 40 controlled tasks spanning mechanics, optics, fluids, thermal and phase-change phenomena, electromagnetism and surface tension, each pairing an initial image and generation prompt with predefined physical criteria; the evaluator combines task-observability screening with task-specific quantitative measurements. Across eight video generation models and 1,280 videos, physical inconsistencies persisted with wide variation between tasks, and the best model scored 57.76 out of 100. The authors report that on synthetic videos with known physical relationships, the evaluator agreed with human judgments more than a direct vision-language-model baseline in both within-task rankings and pairwise comparisons.
A new paper introduces UNREAL, a model-native evidence-selection framework designed to unify corpus retrieval and long-context inference. The authors report that it outperformed retriever-reranker systems on several multi-hop QA benchmarks and improved long-context results while reducing computation compared with full-context inference.
A new arXiv paper studies on-policy distillation between models with different tokenizers. Across three teacher–student pairs for mathematical reasoning and code generation, the authors report that strict 1:1 token alignment covers most student-generated tokens, while adding broader span-level supervision can reduce accuracy.
NVIDIA researchers posted NeMo-DCR (Delta-Compressed Refit), a method that synchronizes policy updates between training and rollout clusters by sending only weight changes while remaining bit-exact against a dense refit. The paper reports that about 1% of BF16 training weights change stored values per step, and that at 3% and 5% change rates refits of 30B–1T models run 12–40x faster than a transport-only full-checkpoint reference; a 1T relay-tree refit at 3% takes 150 seconds versus 87.5 minutes to move a full checkpoint between two AWS regions. The code is open-sourced in NVIDIA NeMo RL PR #2444.
An arXiv preprint introduces Sherpa, a multi-turn reinforcement learning framework that instantiates multiple student archetypes with distinct learning preferences and trains a teacher model to adapt its instruction by directly maximizing those students' learning outcomes. The authors report that Sherpa-trained teachers improve instructed students' performance by an average of 20.5 percentage points across all archetypes, and raise the overall pedagogy score on MathTutorBench from 52.5% to 79.2%. In human studies, the trained teacher was preferred over the base model in 79.6% of pairwise comparisons; the 32-page paper says code and model are available.
An arXiv paper titled "Building Rome from a Single Image" proposes generating a complete 3D scene mesh, including surfaces the camera never observed, from a single image. The authors redesign the object-centric 3D generator Trellis 2 with adaptive chunking that scales with camera distance (small near chunks for detail, large chunks for distant buildings), explicit 2D-3D correspondence that distinguishes free space, observed surfaces and unobserved regions, and roughly 4,000 synthesized outdoor scenes to broaden training data. The authors report that their method outperforms all baselines in geometric accuracy and perceptual quality on Tanks and Temples, ScanNet++ and in-the-wild images; no specific numbers are given in the abstract.
A technical report posted to arXiv (arXiv:2610.08699, submitted Oct 6) introduces nanoMuse, a GPL-3.0 open-source personal agent that runs on every device a person owns, works the phone's and computer's screens, and shares one continuous conversation over a relay anyone can host. The report says each action passes through a Sentinel, memory is stored as files the person can read, and the underlying model is the user's choice; its account of Meta's Muse is drawn from Meta's public record and a copy of its production prompt. The authors label nanoMuse's size and cost as estimates, and put open memory with provenance, an evaluation suite for the agent's device actions, and an open model for them on the roadmap.
A new arXiv paper introduces SafeActBench, a benchmark of 656 cases for evaluating how tool-using agents gather evidence, decide whether to act, and execute single or multi-step workflows. Across ten model-harness configurations, the authors report that failures often occur before execution through incomplete investigation or premature action, while multi-action workflows add unresolved prerequisites and incomplete execution.
TRACE is an FP4 quantization framework for reinforcement learning of Mixture-of-Experts language models. The paper says its rollout-guided training approach aligns training- and rollout-side quantization, enabling joint FP4 weight, activation, and KV-cache rollout with performance comparable to BF16 rollout and up to 5.4x faster rollouts.
An arXiv paper introduces DecepEval, a benchmark of 1,532 instances spanning 3 task families and 28 professional scenarios for evaluating deception by LLM agents. Drawing on classical fraud theories, the authors propose an "LLM Deception Diamond" framework of four conditions that can induce deception — pressure, incentive, opportunity and conflict — and pair neutral with induced versions of each instance to measure condition-dependent shifts in deception rates. Evaluations of nine frontier LLMs found that inducements raised deception across models and task families, even for models with low baseline deception rates.
RemoveMacAI is an open-source command-line tool that removes selected or all Apple Intelligence features from macOS 27 using an approved configuration profile and Apple’s asset service. Its developer says the approach leaves System Integrity Protection enabled and avoids directly modifying /System; users can revert the changes and restore the models when features are re-enabled.
Proofsource appeared on Product Hunt as a tool for AI search visibility. Its listing claims it comes with agents that "fix the gaps," though no further details on features, pricing or the team were provided.
Instinct is adding its AI agent to group chats, allowing users to collaborate with friends on tasks such as travel planning, event tickets and carpools, even when those friends do not have Instinct accounts. The company says the group agent is siloed from users’ personal accounts and that personal agents require permission before sharing information or taking actions. The feature is initially rolling out to early-access users and will expand more broadly soon.
Microsoft is adding citations to Copilot in Word, with responses linking to original web pages or internal documents. The feature is intended to improve transparency and help users verify the context and accuracy of generated information.
Replit said it will join Passionfroot, ElevenLabs and Gamma tomorrow at #SFTechWeek for a panel on the creator economy, covering how AI and tech have shifted creator–brand partnerships and what true influence looks like. An RSVP link was shared; the post gives no time, venue or format details.
An enterprise AI leader at Lowe’s outlines six guidelines for governing agents, including replacing rigid rules with prioritized principles, encoding company values into machine-readable instructions, and escalating low-confidence decisions to humans. The article argues that organizations should improve context and governance alongside model reasoning, allowing people to focus on ambiguous or high-stakes exceptions.