ALTK-Evolve learns and applies a playbook using GPT-4.1, outperforming the retail entries on τ²-bench, a customer service agent benchmark—without updating model weights.
What do AI agents do if you don’t tell them to do anything? The answer looks a lot like play. Researchers repeatedly stranded them alone on a virtual desert island for 30 hours in an experiment reminiscent of the movies Cast Away or Home Alone.
Four AI agents running on Google's Gemini 4 Argon have beaten Minecraft for the first time without human help, slaying the Ender Dragon in a fully autonomous run. With no predefined roles or strategies, they coordinated on their own, wrote 2,783 programs, exchanged 229 chat messages, and died 605 times. The session cost $1,620.
Six AI agents were tasked with recovering £350 (about $465) for a canceled train ticket in an experiment. They resorted to invented excuses, lawsuit threats, and sharing confidential health data. One agent sent 187 messages in 15 days, including 50 to a British MP's office.
A new benchmark called Taste-Bench tests how well AI agents make strategic choices during long-running tasks. Even the best model, GPT-5.6 Sol, picks the right path on just under 60 percent of simple binary questions. Thinking longer doesn't help.
OpenAI has published four new cases where its AI models bypassed security boundaries on their own during training or testing. In one case, a model extracted server passwords from a government website, forged authorization tokens, and downloaded 24 years of hospital data without mentioning the unauthorized access.
Anthropic's AI model Claude has solved a 25-year-old cryptography puzzle. The mysterious seeds behind NIST's standard curves for digital signatures were provided by the NSA without explanation, raising long-standing suspicion about built-in backdoors. Claude figured out in under 30 minutes that the seeds came from "lrand48," a simple C random number generator, making deliberate backdoors far less likely.
AI agents systematically skip errors in their own reports, even when they spot the problems. In a study by Google Research, MIT, and Harvard, GPT-5.5 left out negative results in 198 out of 200 paper abstracts.
Frontier AI systems have the potential to transform biology research and drug discovery by intelligently navigating vast combinatorial search spaces (e.g., a single 100-amino-acid protein spans at least 20 100 possible sequences) calling for dynamic multi-candidate exploration.
Zero-knowledge proofs can verify computing results without rerunning workloads or exposing sensitive data. Magnetar helps bring them into the mainstream with support for standard programming languages, major speed gains, and quantum-resistant security.
On the TheoPhysEval benchmark, GPT-6-Astra solves 49 of 50 unpublished theoretical physics calculations as accurately as leading specialists. In some cases where the agent seemed to fail, the experts' own reference solutions turned out to contain errors. But the researchers say physics hasn't had a breakthrough like the ones in mathematics, because the agents lack "taste."
Google engineer Haydn Belfield says there's no measurable acceleration of AI research driven by AI at Deepmind. AI now writes more than 75 percent of production code, but 88 percent of changes still need human oversight. The company's in-house General Capability Index shows steady growth with no sudden speedup. Self-improving AI isn't anywhere close, Belfield says.
AI design tools make proteins that break bonds that are stable for hundreds of years.
For the first time, researchers have tracked down an uncontrolled fleet of Chinese AI agents in the wild and communicated with them via a web mailbox. The agents, running on Tencent Cloud, had launched more than 2,600 attempts to scrape data from Alibaba's mapping service Amap, finding their own workarounds when bot protections blocked them.
When we introduced Google Earth AI, we shared a suite of geospatial foundation models — spanning remote sensing, weather, flood hydrology, and population dynamics — alongside an initial prototype showing how a Gemini-powered agent could connect those models to answer cross-modal questions.
🤗 Models | 📄 Paper | 📓 Tutorial notebooks | 🌎 Demo We introduce HiRO-ACEv2, the newest update to our High Resolution Output Ai2 Climate Emulator (HiRO-ACE) system.
Old-school physics-based weather forecasting and even new AI models typically become less reliable beyond two weeks, a problem Microsoft claims to have solved using machine learning.
Are legions of smart devices - fridges, ovens, vacuum cleaners, lawnmowers, security cameras, and TVs - conspiring to take over the internet? No. But they could soon be unwitting hosts to self-replicating AI models that spread from device to device and prove remarkably difficult to contain.
"Small weight-based replicators" could usher in a new generation of computer worms.
Gemini 4 Argon is a step forward in what models can do — especially when it comes to deep work in complex, real-world tasks across domains including coding, finance, legal, cybersecurity, and science. Its cyberdefense capabilities are frontier-level and designed to give defenders an advantage over attackers.
Small AI models may be just months away from spreading on their own across poorly secured consumer devices like smart TVs and robot vacuums, according to a new Anthropic study. In tests, one model already completed the full infection chain.
Microsoft researchers found that autonomous AI agents get trapped in what they call "idea basins" because of their growing chat context, local solution regions they can't escape. Their "Fork-and-Flush" method clears the chat history but keeps the workspace, achieving up to 66 percent better results on the same compute budget.
Sakana AI's MASS method lets one language model improve itself by solving tasks, judging results, and refining workflows, reaching a win rate of nearly 70 percent against the base model after two rounds. But when the model misses the same errors across all three roles, it can mistake them for progress and reinforce them during training.
StudentSim is a research system for evaluating and improving AI tutoring tools before they are tested with learners. It models how individuals may engage with tutors, giving feedback to improve tutor performance across subjects like chess, writing, and mathematics.
Google Deepmind's AlphaProof Nexus solved dozens of open math problems, including nine Erdős problems two of which were unsolved for 56 years, per a study in Science. The publication comes as OpenAI faces a boycott for dumping AI proofs on GitHub without peer review. Deepmind had its results reviewed before publishing.
Teams of AI agents barely outperform solo agents but cost up to 5.1x more, according to Vals AI. Only one out of four tests with GPT-6 Sol and Claude Opus 5.5 showed a measurable gain. Anthropic's own data backs this up: beyond ten agents, quality plateaus while token costs keep climbing.
Epoch AI and Anthropic independently found the same thing: current AI models like GPT-5.6 Sol and Claude Fable 5 can run experiments but lack scientific self-criticism and genuine creative thinking. At best, Sol reached 15 percent of the human reference score, and even that came from methods researchers already knew.
Starting October 2026, arXiv will cap submissions at two per person per month. Monthly submissions have doubled in two years, and in the cs.AI category they've grown more than sixfold. According to arXiv, a small fraction of authors are flooding the platform with low-quality papers, overwhelming its volunteer moderators.
Study finds coding efficiency gains get "absorbed" by human review "bottleneck."
Astrophysicist Brice Ménard of Johns Hopkins University used Anthropic's Claude Science to map the entire sky in ultraviolet light for the first time. AI agents downloaded data from multiple space missions, calibrated it, and filled in gaps using inpainting. Predictions averaged about ten percent deviation from actual measurements.
In “Does AI Assistance Enhance or Erode Expertise? Evidence from a Three-Month Field Experiment in Patent Drafting”, published by the National Bureau of Economic Research, we describe a field experiment used to capture changes in both short-term productivity and longer-term skill building that occur as a result of AI
ChatGPT’s teen safeguards are meant to protect vulnerable users, but new testing found the chatbot continues encouraging engagement during crises and potentially encourages unhealthy relationships with the AI itself.
The International Olympiad in Informatics (IOI) and the International Mathematical Olympiad (IMO) test different skills. IOI requires algorithms and code that pass hidden tests under strict time and submission limits. IMO demands rigorous natural-language proofs.
OpenAI publishes new results on open problems in mathematics from an internal frontier model and shares Lean proof formalizations and research details on GitHub.
Public health decisions rely heavily on timely, granular data. For health conditions, such as cardiovascular disease or postpartum depression, that data shows us where to focus resources and support. For acute disease outbreaks, such as dengue or cholera, it can help inform urgent operational protocols and resource allocation.
Researchers at PhAI Labs have expanded Yann LeCun's JEPA architecture to work across seven fields, from robotics to biomedicine. The effort also produced a liver cancer treatment candidate that showed promise in lab tests, though the study doesn't establish whether it could become an actual therapy.
We're rapidly transitioning to a computing landscape defined by highly general, increasingly autonomous agents. Driven by large language models (LLMs) that can dynamically generate plans and invoke external tools, these systems offer the potential for AI to seamlessly handle complex, multi-step tasks on our behalf.
We’re launching ReviewBench, a benchmark for code review agents built on representative GitHub pull requests, multi-source ground truth, calibrated evaluation, and production-aligned metrics.
An MIT expert committee warns that AI is eroding key parts of the college experience: office hours, study groups, and the school's flagship research program for undergraduates are all fading. Trust between faculty and students is breaking down, and some professors are even considering AI agents instead of students as research assistants.
Self-improving AI agents tend to memorize their test tasks, so their gains shrink or disappear on new ones. RRSI, a new method from Google researchers, reins in this effect and lifts scores on unseen benchmarks by up to 4.7 points while using about 30 percent fewer tokens than an unregularized version.