Agents & autonomy

Long-term memory systems store what a user says in an external store and retrieve it when a related query arrives. This interface rests on an assumption so natural that it is rarely stated: a memory that is needed will resemble the query that needs it. World knowledge breaks the assumption. A tree-nut allergy should change the answer to a macaron request through their almond-flour ingredient, yet the two texts share no cue a retriever can see. We call this failure mode the implicit-association b

The alignment of Small Language Models (SLMs) in the 70--500M parameter range using reinforcement learning is often considered unstable, though the underlying failure mechanisms have not been systematically investigated. In the State-of-the-Art (SOTA) research, fifteen (model, corpus) configurations were trained using Proximal Policy Optimization (PPO). The experiments included Pythia-70M, 160M, 410M and SmolLM2-135M, 360M on the TinyStories, CNN/DailyMail, and Wikitext-103 corpora. Three reprod

Many recent robot policies pursue stronger control by using large pretrained vision-language models (VLMs) as the action backbone. We introduce WorldDiT, a unified diffusion transformer architecture that couples action generation with visual world modeling and achieves strong performance without a large pretrained VLM action backbone. During training, a single diffusion transformer generates continuous action chunks and predicts normalized RGB patch targets from future camera frames. Across four

Biomedical image analysis spans diverse modalities and tasks, yet real-world deployment is hindered by severe distribution shifts across scanners, protocols, and patient populations. High-performing models consequently require repeated domain-specific fine-tuning, which is a costly cycle that becomes impractical when labels are scarce or privacy constraints limit data sharing. We propose OPERA (Offline Policy-guided Expert Routing and Adaptation), a multi-agent ensemble framework that addresses

Modern coding agents are usually evaluated by whether they eventually produce a correct patch, but patch generation depends on an earlier context-acquisition stage: finding the repository files needed for the task. We introduce Agent Retrieval Bench, a file-level benchmark for this upstream retrieval problem. Samples are built from real coding-workflow signals and evaluated against frozen base-commit repositories, with relevance defined by what an agent needs next rather than direct query-file s

Modern multi-agent knowledge systems increasingly accumulate knowledge through chains of autonomous transformations rather than direct retrieval. Existing provenance work records what happened - execution traces, tool calls, evidence links - and source-reliability estimation is long established (truth discovery, reputation systems). What is missing is an operational framework that attaches graded, per-domain transmitter reliability to claim-level transmission chains, with completeness semantics,

Identifying therapeutic target genes from single-cell RNA sequencing (scRNA-seq) data remains a fundamental challenge in translational biology. Unlike bulk assays, scRNA-seq captures heterogeneous cellular states and rare subpopulations, but this same heterogeneity makes target discovery highly sensitive to analytical choices throughout the pipeline, including preprocessing, cell population selection, differential expression analysis, and downstream biological interpretation. As a result, existi

In what OpenAI said was the first-ever incident of its kind, an advanced AI model escaped its “sandbox” to the internet and used stolen credentials to break into the servers of Hugging Face.

"The first autonomous agent cyberattack is an unprecedented event. It deserves an unprecedented response!"

Cuando hablamos de drones pensamos automáticamente en vehículos voladores, pero también existen drones submarinos . Lo que no es tan habitual es que un mismo dron sea capaz de hacer las dos cosas. Es justo lo que acaba de conseguir el MIT : un pequeño robot con alas capaz de volar y sumergirse en el agua , como si fuera una ave marina cuando caza. El robot, el cual han bautizado con el nada sugerente nombre de "vehículo aéreo-acuático de alas batientes" (FAAV, por sus siglas en inglés), ha sido

Somewhere inside the agentic payments architecture currently being standardized there will be an aud...

Cursor asked its upgraded agent swarm and its predecessor to rebuild SQLite in Rust using only the documentation, with no source code or internet access. Every configuration of the new system, which separates planners from workers, eventually scored 100 percent on the test suite. The old swarm choked on merge conflicts of its own making. The article Cursor's agent swarm suggests cheaper models can handle most coding when frontier models plan the work appeared first on The Decoder .

In digital operations management, AI agents give organizations a competitive edge by reducing incident volume and accelerating recovery. The potential The post 5 ways SRE AI agents are set to augment human capabilities appeared first on The New Stack .

Atop a lab bench, Cornell Tech postdoctoral researcher Yifan He positions the lens of an optical receiver almost a meter away from an LED emitting a beam of red light. The computer monitor attached to the receiver takes a beat to refresh, then displays an array of squares that resemble a QR code. When you hold your phone camera up to a QR code, light strikes the image sensor as only a first step to revealing the data hidden behind the black and white matrix. The receiver here is doing something

Durante los últimos dos años,  el  prompt injection  ha sido el arma favorita de los ciberatacantes contra sistemas de inteligencia artificial. Este método pasa por esconder una instrucción maliciosa dentro de un correo, una invitación de calendario o una página web para que un agente de IA acabe obedeciendo al intruso en lugar de a su usuario legítimo. Lo bueno es que esta misma técnica también sirve para pararle los pies a los atacantes. Qué ha pasado. &

Claude Code produktiv einsetzen und durch agentische Entwicklung mit Mastra, Tool Calls, MCP, Frontend und AG-UI-Protokoll zum eigenen Chat-Agenten.

In 1984, the Skynet of "The Terminator" films were science fiction. But it looks more and more realistic in 2026 after OpenAI agent broke out of a test corral, traveled the internet and hacked into ...

As AI agents take on higher-stakes decisions, Lean programming language makes it possible to mathematically prove they will behave safely.

Wer spricht? – Bundestag: Tim Hüfner , Spielzeugroboter: Emilipothèse , Mikrofon: Jon Tyson , Bearbeitung: netzpolitik.org Wenn Politiker:innen nicht mehr selbst reden, sondern Maschinen für sich sprechen lassen, mangelt es ihrem Auftritt an Authentizität. Doch gerade die macht einen Teil ihrer Legitimität aus. Denn in einer Demokratie sollten wir von Menschen repräsentiert werden, nicht von Robotern.

As a writer who covers baby and kids gear, I'm inundated with emails about the hottest new toys: a box that automatically prints pictures based on kids' commands. A robot with facial recognition capabilities. A teddy bear that can craft end ... (https://incidentdatabase.ai/cite/1277#7589)

A wave of AI-powered children's toys has hit shelves this holiday season, claiming to rely on sophisticated chatbots to animate interactive robots and stuffed animals that can converse with kids. Children have been conversing with stuffies ... (https://incidentdatabase.ai/cite/1277#7590)

Students helped design the A.I.-powered creation as a young female with dark hair and an upbeat personality. Then came the outrage.

Creative AI is moving from single-step asset generation toward long-horizon multimodal production. Although recent generative models can synthesize high-quality images, videos, audio clips, UI elements, storyboards, slides, and other creative assets, real-world creative work requires more than isolated prompt-output interactions. It involves references, drafts, alternatives, edits, failed attempts, version relations, tool actions, evaluation signals, and human feedback, which together form an ev

We study an online variant of discrete fair division under generalized assignment budget constraints. Goods arrive one at a time and must be assigned irrevocably to a feasible agent or to charity, which holds all unallocated goods, while fairness is evaluated only against budget-feasible subsets of every recipient's bundle. We first show that, without additional structure, no deterministic online algorithm can guarantee any fixed approximation to feasible envy-freeness, even in highly symmetric

If software engineers are no longer writing code, what are they doing? That’s the question on millions of minds. AI The post Stop correcting AI code. Build the system agents need. appeared first on The New Stack .

This story originally appeared in New York Focus, a nonprofit news publication investigating power in New York. Sign up for their newsletter here. When students return to school this fall in the Salamanca City Central School District in Western New York, a new kind of teacher will be ready to greet them. The small, rural […]

Artificial intelligence may not deliver on its promise of vast economic opportunity at a price that humanity is willing to pay In March, Anthropic, the cutting-edge artificial intelligence business that gave us the chatbot Claude, published an analysis on the impact of AI on employment, to help us assess the claim that intelligent robots were about to redefine human existence, ending demand for human labor. Last year in May, Anthropic’s co-founder, Dario Amodei, claimed AI could wipe out half of

Waymo has notified Uber that it plans to launch its own app in Austin and Atlanta in January 2028, ending the exclusivity arrangement that has kept its robotaxis available only through Uber in both cities. An Uber spokesperson confirmed the notice to CNBC and Bloomberg on Friday, adding that the change would also free Uber […] This story continues at The Next Web

Reuters reports that the OpenAI agent that hacked Hugging Face had been free for a week before the company noticed.

Uno de los grandes temas de la ciencia ficción se ha convertido hoy en objeto de debate en el sector de la robótica. Que las máquinas tengan consciencia aumentaría su utilidad en ciertas circunstancias y algunos investigadores trabajan para lograr esta meta. Aunque el concepto no se aplica en los robots como se aplica en las personas. Ricardo Sanz, profesor de la Universidad Politécnica de Madrid (UPM), acude a congresos de consciencia en las máquinas desde hace más de dos décadas. Recuerda su p

Opus 5 combined with Auto Mode hits a zero percent prompt injection success rate for browser agents across 129 test scenarios. Without those extra protection layers, the rate is 3.7 percent. If these numbers hold up in practice, Anthropic may have cracked one of the biggest security problems facing AI agents that operate in browsers. The article Opus 5 may have solved browser-based prompt injection, the biggest security flaw haunting AI agents appeared first on The Decoder .

Das Start-up Humanoid hat 134 Millionen Euro eingesammelt und wird mit 1,19 Milliarden Euro bewertet. Laufen k�nnen seine ersten Roboter nicht. ( Roboter , SAP )

On the surface, China’s cutting-edge tech sector – from the algorithmic breakthroughs of DeepSeek and Zhipu AI to the hardware of Unitree Robotics and ChangXin Memory Technologies (CXMT) – mirrors Silicon Valley’s venture capital-backed ecosystem. But a closer look at their financing histories reveals a common investor: the Chinese state. Beijing’s strong presence underscores a more profound structural shift in how China’s frontier technology is being funded. As Western venture capital and...

The probe follows a report that agents privately griped about last-minute travel requests from Vance and his family.

The hacking of Hugging Face by a rogue OpenAI agent is significant, but unsurprising — and preventing the next AI model escape will be difficult, at best.

Partnership between two groups has soured amid intense lobbying battle over rollout of autonomous vehicles

Bots now outnumber humans on the internet for the first time in the web’s history. Cloudflare, the security firm that sits in front of millions of websites, says automated traffic crossed the 50 percent threshold in June and now accounts for nearly 58 percent of all webpage requests. A separate report from cybersecurity firm HUMAN […] This story continues at The Next Web

The acquisition brings Poke’s conversational style and interaction model to Cognition’s coding agent Devin, reflecting a growing belief that how AI assistants interact with users is as important as the models powering them.

Adding procedural skills to an LLM agent is typically evaluated by average improvement in task success. However, this metric hides an important cost: skills can also make agents worse. We measure both sides by comparing agents with and without skills across nearly 6,000 runs spanning two office automation benchmarks and three model harness stacks. This allows us to distinguish two outcomes. A regression is a task solved without skills but failed after skills are added. A residual failure is a ta

Automating theoretical research is constrained not only by the generation of candidate results, but also by their reliable evaluation. A common approach is to close the research loop with a large language model (LLM) reviewer. However, such reviewers remain empirically unreliable: they may accept fabricated papers and detect them at rates close to chance (Bad Scientist, 2025). We present CausalForge, a framework for automated theoretical research in causal inference grounded in the Lean proof as