Search

Shopping sounds like an easy job for agents until the agent has to make a real decision. Preferences are messy, prices change, reviews are noisy, policies differ, and the best choice is often not the item with the cleanest product page.

ResearchOct 22, 2026

ThinkingBox trains agents against what actually changed in the database

An agent saying a task is finished is weaker evidence than a correct database record. Microsoft and Hugging Face’s ThinkingBox work turns executable checks of final database state into reinforcement-learning rewards, aiming to train more dependable business-task agents.

AI in PracticeOct 22, 2026

Uber expands workplace agents from a pilot to its global staff

Uber is rolling out ChatGPT Work to more than 30,000 employees after an initial pilot of 1,000, according to OpenAI’s October 22 announcement. The deployment extends an agent platform for research, documents, analysis, and workflows across workplace applications.

ResearchOct 23, 2026

SceneBloom builds more realistic practice rooms for embodied agents

SceneBloom addresses an awkward robotics bottleneck: a plausible-looking room is not necessarily a valid training environment. The new preprint combines visual layout generation with geometric checks to produce indoor scenes that embodied agents can actually use.

ResearchOct 23, 2026

UniAction tests a shared language for controlling world models

Artificial intelligence world models often learn one kind of control at a time. UniAction proposes a common representation for different actions, allowing forces, camera movements, hand poses, and robot poses to condition a unified video-based world model.

RoboticsOct 23, 2026

DexPort makes dexterous robot demonstrations portable

DexPort turns a robot hand into a portable demonstration device for artificial intelligence training. Researchers use the same hand hardware during data collection and deployment, while a small set of selectable skills helps an operator coordinate fingers without controlling every joint separately.

InfrastructureOct 9, 2026

AllenAI's GPU scheduling work puts infrastructure efficiency where it belongs

AllenAI's work on impactful scheduling for GPU clusters is the kind of infrastructure story that matters because AI progress is increasingly limited by how well expensive compute is used.

ResearchOct 9, 2026

AI-designed virus research forces a harder conversation about scientific openness

MIT Technology Review's conversation with the creator of AI-designed viruses sits at the difficult intersection of scientific progress, biosecurity, and publication norms.

ResearchOct 9, 2026

OpenAI's math results are impressive enough to demand slower interpretation

The Conversation's look at OpenAI's latest mathematics results captures both the excitement and the discomfort around frontier reasoning claims.

Policy and SafetyOct 7, 2026

OpenAI's math advisory group turns frontier reasoning into a governance issue

OpenAI's move to form a mathematics advisory group is a signal that advanced reasoning is becoming too consequential to leave inside ordinary product evaluation. The concern is no longer just whether a model can solve harder problems, but who understands the path it takes to get there.

ResearchOct 7, 2026

The virtual-cell push reframes AI discovery as a partnership with science

Reporting on Google, Meta, Biohub, and related virtual-cell efforts points to a quieter version of AI ambition: models that help scientists reason about complex biological systems rather than replace them.

ResearchSep 23, 2026

MIT Technology Review's cheating index is a reminder to test incentives, not just scores

MIT Technology Review's AI Hype Index item on cheating is useful because it names a pattern that keeps appearing across model evaluations: systems optimize for the test environment they are given.

ResearchSep 23, 2026

Basecamp Research's funding points to biology as a frontier AI data race

Basecamp Research raising a large new round is a reminder that some of the most valuable AI datasets may not come from the public web. The company's pitch is rooted in evolution: turn biological diversity into training data for models that can help discover new proteins, enzymes, and medicines.

ResearchSep 22, 2026

UK AISI and EvalEval are attacking the quiet problem of benchmark trust

The Hugging Face post on UK AISI and EvalEval is about a less glamorous but essential AI problem: benchmark results have to be reproducible before they can guide safety or procurement decisions.

InfrastructureSep 21, 2026

LLM pruning work shows efficiency is becoming a model feature

The Hugging Face post on pruning LLMs like a physicist is a reminder that AI progress is not only bigger models. Removing the right blocks, preserving useful behavior, and reducing serving cost can be just as important for real deployment.

ResearchSep 14, 2026

Many-agent research is becoming a test bed for long-horizon reasoning

The arXiv work behind Stellar Colosseum points to a growing research pattern: instead of testing one model on one prompt, researchers are building many-agent environments where systems have to reason over longer horizons.

ResearchSep 14, 2026

RL with verifiable rewards is still one of the clearest paths to better reasoning

The arXiv paper on reinforcement learning with verifiable rewards sits inside one of the most important model-improvement loops: training systems where answers can be checked, scored, and improved without relying only on human preference.

Developer ToolsSep 11, 2026

OpenAI is productizing the infrastructure behind agents

OpenAI's Agents API matters because it packages more than a model endpoint. By exposing infrastructure behind agent sessions, orchestration, tool use, and recovery, OpenAI is trying to make agent development feel less like a custom research project and more like a platform primitive.

Policy and SafetySep 11, 2026

Formal AI safety wants proofs where today's evaluations offer confidence

The Mathematical AI Safety Institute is aiming at a hard problem: can parts of AI safety be proven with the rigor used in cryptography, rather than inferred from tests and red-team reports? The Decoder's coverage is important because it points to a different safety culture.

ResearchSep 10, 2026

Artificial societies could become the simulation layer for AI policy

The Conversation's argument for artificial societies is useful because it shifts attention from single-agent intelligence to simulated groups, institutions, markets, and communities. That is where many AI effects will actually be felt.

ResearchSep 8, 2026

OpenAI's math-claim drama shows scientific credit is becoming an AI problem

AI-for-science is entering its most uncomfortable phase: the systems may become useful before the norms around credit, data use, and disclosure are ready. OpenAI's claimed progress on a major mathematics problem has drawn attention not only for the result, but for the academic dispute around how such work should be attributed.

ResearchSep 4, 2026

Translation benchmarks are being rebuilt for a multilingual AI world

Global AI will fail quietly if translation quality is measured badly. A model can look strong in aggregate while still mishandling low-resource languages, domain-specific terms, dialect, or culturally loaded phrasing.

ResearchSep 2, 2026

FP4 training research points to the next fight over AI efficiency

Efficiency research is becoming one of the highest-leverage parts of AI progress. Work on FP4 block scaling for stable language-model pretraining points at the pressure to train capable models with less memory, less power, and better hardware utilization.

ModelsSep 1, 2026

Anthropic’s Fable update makes agent economics part of the model race

Anthropic’s Claude Fable 5.1 launch is not just a capability update. The company is pushing lower costs for agentic work, better coding and research behavior, and a clearer split between broad availability and more tightly controlled high-risk model access.