Google AI Introduces EnvHarness: A Programmable Layer That Turns Static Agent Environments Into Adaptive Training Worlds
A team of researchers from Google Cloud AI Research, Washington University in St. Louis and UNC Chapel Hill has released EnvHarness, a programmable layer that turns a static agent benchmark into one that adapts to the policy training on it. LLM agents now learn less from curated text and more from interactive environments, but those environments are hand-built and frozen: they behave identically no matter which agent is acting or how much it has improved. The usual fix is to generate new enviro
EnvHarness is a programmable layer that modifies static training environments for AI agents by wrapping them with plug-in components that change where episodes start, what actions agents can take, and what they observe, while keeping the original human-built verification system intact. The tool matters because AI agents currently learn from hand-built environments that behave identically regardless of agent performance, limiting their ability to improve, and generating new environments requires domain-specific pipelines and unreliable automated verifiers. EnvRigger, an LLM component of the system, automatically diagnoses weaknesses in agent performance from rollout observation and writes targeted environment modifications, which are then validated on fresh rollouts. Across five benchmarks, skills trained in these adapted environments showed gains of up to 9.0 points on held-out tasks while requiring 9.8% fewer execution steps than skills trained in original environments.
Explainer: How to Actually Read an AI Benchmark (Without Being Fooled)
One short email a day with the stories that matter. Free. Unsubscribe with one click.
Want to go deeper than the news? Explore live, cohort-based AI courses taught by practitioners.
Browse AI courses on Maven
OpenAI’s quarterly revenue reportedly crosses $10B for the first time
Sales of $10B represents an almost 50% jump from revenue of $6.7 billion the company recorded in the second quarter.
Agents that build their own databases are forcing a data foundation rethink
AI agents are changing how enterprises create, move and govern information, and data observability is turning into a prerequisite rather than a nice-to-have. Models get most of the attention, but the data beneath them decides whether they hold up in production. That pressure is now reaching the cloud-native community, where KubeCon organizers added a dedicated […] The post Agents that build their own databases are forcing a data foundation rethink appeared first on SiliconANGLE.
Hands-on with Ring’s new AI drone camera for your home
Ring’s new Drone Cam is a flying indoor security camera. Ring is finally shipping a flying indoor drone camera. For real this time. The Ring Drone Cam will arrive on January 27th, 2027, for $249.99, Ring founder Jamie Siminoff said at a launch event in New York City this week. If it does actually ship, it will arrive almost six years after the company announced its first flying camera, the Ring Always Home Cam. Designed to patrol inside your home when you're not there, the Drone
Also reported by TechCrunch and Ars Technica
OpenAI rolls out new business profiles, making ChatGPT just a little bit more like Google
OpenAI is looking to make ChatGPT a more robust consumer discovery engine, and a host of new features seek to empower businesses to help it do that.
Also reported by The Verge
AI agents carry memory and storage like people, so their data needs guarding while in use
As AI agents move into production on Kubernetes, cloud engineering teams at financial companies face a security problem that encryption alone can’t solve. Container orchestration has become the default foundation for AI workloads, and the data those agents use is now the main target. Payment networks feel that pressure more than most because every transaction […] The post AI agents carry memory and storage like people, so their data needs guarding while in use appeared first on Silic
Serve Robotics opens robot delivery network for others to build on
The AV Delivery Developer Portal enables anyone to call robots from Serve’s delivery fleet. Source: Serve Robotics Serve Robotics Inc. today launched the AV Delivery Developer Portal, which it said gives developers, businesses, hobbyists, and AI assistants the opportunity to programmatically summon its sidewalk delivery robots. The company said its public application programming interface, or API, allows others to work with its autonomous robot fleet. It also released a sandbox environment