Research

I signed up for another SaaS

new software benchmark…

Anthropic study shows AI needs hours, not weeks, to build exploits from security patches

Anthropic's security team found that its Mythos Preview AI model can turn security patches for Firefox and the Windows kernel into working exploits within hours…

Quantum’s breakthrough

Quantum is approaching an inflection point, bringing significant economic opportunity—and potential security risks. Are business leaders ready to harness quantu…

Introducing Google AI Threat Defense to help you outpace the adversary

aside_block <ListValue: [StructValue([('title', 'Summary of today’s news'), ('body', <wagtail.rich_text.RichText object at 0…

An update on US consumer sentiment: Gloomier outlook ahead of sunnier days

Ahead of the summer season, a smaller share of US consumers reported optimism while a greater share reported cost concerns, which could impact their spending ac…

Evolving Dataflow to process massive datasets for machine learning

Google created MapReduce more than 20 years ago to solve the scaling problems in data processing that the then young company was running into. The AI era that w…

How Trustpilot built a real-time architecture for data enrichment using Gemma

Processing millions of user reviews in real-time, under strict latency and cost constraints, is no easy task. Trustpilot has been doing exactly that with custom…

Latino economic mobility: Insights from five years of research

This compendium brings together five years of McKinsey research on how Latinos are reshaping America’s economic future and why expanding opportunity for this dy…

Automate model quota request and operational issue triage on Amazon Bedrock

In this post, we introduce Amazon Bedrock Ops Alert, a three-layer automated monitoring solution that proactively detects operational issues, dynamically adjust…

Resetting cost competitiveness in pulp and paper packaging

Overcapacity, volatile costs, and weak demand are compressing margins. Successful companies are harnessing AI to pursue cross-functional interventions that targ…

NVIDIA Nemotron 3 Ultra now available on Amazon SageMaker JumpStart

Deploy NVIDIA Nemotron 3 Ultra on Amazon SageMaker JumpStart. Get 5x faster inference and 30% lower cost for agentic AI workloads with this frontier reasoning m…

HR Monitor 2026: A turning point for the people function

A comprehensive benchmark survey of workforce and HR trends across Europe, the United States, and China delivers critical insights for today’s HR leaders.…

How to unlock true ROI in software development – a deep dive into the latest DORA research

How do you prove the business value of generative AI to your teams? Technology and finance leaders need to show the clear business value of AI projects to secur…

Scale Robot Reinforcement Learning with NVIDIA Isaac Lab on Amazon SageMaker AI

In this post, we show how to train robot policies for the Unitree H1 humanoid with NVIDIA Isaac Lab on Amazon SageMaker AI across two compute options: Amazon Sa…

Optimize blueprint extraction accuracy in Amazon Bedrock Data Automation

Blueprint instruction optimization is a BDA feature that automatically refines your extraction instructions to address this challenge directly. You provide thre…

Evaluate AI agents systematically with Agent-EvalKit

Agent-EvalKit is an open-source toolkit (Apache 2.0) that makes this evaluation infrastructure available by integrating with AI coding assistants, including Cla…

Fine-tune FLUX.1 with your own images

We've added fine-tuning (LoRA) support to FLUX.1 image generation models. You can train FLUX.1 on your own images with one line of code using Replicate's API.…

Fine-tune FLUX.1 to create images of yourself

Create your own fine-tuned Flux model to generate new images of yourself.…

Fine-tune FLUX.1 with an API

Create and run your own fine-tuned Flux models programmatically using Replicate's HTTP API.…

Using synthetic training data to improve Flux finetunes

It's easy to fine-tune Flux, but sometimes you need to do a little more work to get the best results. This post covers techniques you can use to improve your fi…

Made in Figma: The National Park Service goes from paper to pixels

You can now fine-tune open-source video models

Train your own versions of Tencent's HunyuanVideo for style, motion, and characters on Replicate.…

Introducing Stable Virtual Camera: Multi-View Video Generation with 3D Camera Control

Introducing Stable Virtual Camera, currently in research preview. This multi-view diffusion model transforms 2D images into immersive 3D videos with realistic d…

IDC study says the global workforce engaged in software design is expanding

Introducing EVMbench

OpenAI and Paradigm introduce EVMbench, a benchmark evaluating AI agents’ ability to detect, patch, and exploit high-severity smart contract vulnerabilities.…

Advancing independent research on AI alignment

OpenAI commits $7.5M to The Alignment Project to fund independent AI alignment research, strengthening global efforts to address AGI safety and security risks.…

Our First Proof submissions

We share our AI model’s proof attempts for the First Proof math challenge, testing research-grade reasoning on expert-level problems.…

Why we no longer evaluate SWE-bench Verified

SWE-bench Verified is increasingly contaminated and mismeasures frontier coding progress. Our analysis shows flawed tests and training leakage. We recommend SWE…

Pacific Northwest National Laboratory and OpenAI partner to accelerate federal permitting

OpenAI and Pacific Northwest National Laboratory introduce DraftNEPABench, a new benchmark evaluating how AI coding agents can accelerate federal permitting—sho…

Joint Statement from OpenAI and Microsoft

Microsoft and OpenAI continue to work closely across research, engineering, and product development, building on years of deep collaboration and shared success.…