Blog

Scale Labs is Scale’s new research hub studying how advanced AI systems behave in real-world environments. Building on the work of SEAL, the lab focuses on evaluation, agentic and multimodal systems, post-training methods, enterprise deployment, and AI risk and oversight infrastructure, with research spanning frontier capability measurement, long-horizon agent behavior, and national-scale safety evaluation.

Read more

As LLMs and coding agents take on professional coding work, evaluations must assess them like junior engineers: by how they investigate a system, gather evidence, and explain what they’re observing. SWE-Atlas, the first evaluation suite of its kind, does just that.

Read more

Scale RL Environments provide simulated, high-fidelity systems where AI agents can learn through trial and error across real workflows. By pairing realistic data universes with structured trajectories and process-level verification, teams can train, evaluate, and iterate on agent behavior safely, producing measurable improvements on public benchmarks while integrating directly into existing agent pipelines.

Read more

Scale AI partners with Webster University to launch a technical writing certificate advancing AI workforce skills.

Read more

The emergence of Moltbook offer an early glimpse into AI agents operating collectively at scale. These systems demonstrate how interaction between autonomous agents can give rise to emergent behaviors that are not attributable to any single model or prompt, complicating traditional, model-level approaches to AI safety. Viewed through the AI Risk Matrix, such agent collectives point to a distinct and underdeveloped category of system-level risk with implications for evaluation, red teaming, and governance.

Read more

Scott O’Neill is a plumbing sales professional in Louisiana who contributes to building better AI models in his spare time. Between a full-time job and raising two young daughters, Scott uses flexible, remote work through Outlier to stay connected to technology, apply his problem-solving skills, and continue learning. His story reflects how people from diverse backgrounds are helping shape the future of AI on schedules that fit real life.

Read more

Scale CEO Jason Droege reflects on a record-breaking 2025 and shares how Scale is building reliable, production-ready AI systems for 2026.

Read more

What it will take for the United States to move from AI experimentation to real governance, government-wide implementation, and lasting global leadership.

Read more

The team is kicking off 2026 like the rest of us: by focusing on health(care)! They discuss why adopting AI in healthcare is different from other enterprise AI initiatives and how leaders can account for those differences. And as always, they react to some of the internet's hottest takes on AI (healthcare edition).

Read more