What’s Next in AI
is foundation models at scale
The AI landscape today is dominated by purpose-built models deployed for dedicated tasks. But enterprises need a large corpus of labeled data, significant resources, and teams of skilled data scientists to train and maintain these models. Foundation models represent a generational opportunity for enterprise. They’re general-purpose, pre-trained models that can be fine-tuned to accomplish a wide set of tasks. We’re developing software, middleware, and hardware to bring frictionless, cloud-native development and use of foundation models to enterprise AI.
Introducing watsonx.ai
Explore our next-generation enterprise platform, powered by IBM's full technology stack and designed to enable enterprises to train, tune, and deploy AI models.
Our work
Everything IBM Research showed off at Think 2023
NewsMike Murphy and Kim MartineauHow Intel oneAPI tools are accelerating IBM's Watson Natural Language Processing Library
Technical noteStephanie Kuan, Deb Bharadwaj, Preethi Venkatesh, Shankar Ratneshwaran, and Waleed KhanBuilding a foundation for the future of AI models
NewsMike MurphyA cloud-native, open-source stack for accelerating foundation model innovation
Technical noteTalia Gershon, Priya Nagpurkar, Carlos Costa, and Darrell ReimerEarth’s climate is changing. IBM’s new geospatial foundation model could help track and adapt to a new landscape
NewsSriram Raghavan and Christina ShimIBM and PyTorch change one line of code to massively improve AI model training
ReleaseMike Murphy- See more of our work on AI
Tools + code
IBM Analog Hardware Acceleration Kit
An open source Python toolkit for exploring and using the capabilities of in-memory computing devices in the context of artificial intelligence.
View project →Project Debater for Academic Use
The technologies underlying Project Debater available as cloud services. Includes core natural language understanding capabilities, argument mining, and narrative generation.
View project →AI FactSheets 360
Toolkit to create factsheets outlining the details about how an AI service operates, how it was trained and tested, its performance metrics, fairness and robustness checks, intended uses, maintenance, and other critical details.
View project →GT4SD
An open-source library to accelerate hypothesis generation in the scientific discovery process.
View project →
MIT-IBM Watson AI Lab
We’re partnering with the sharpest minds at MIT to advance AI research in areas like healthcare, security, and finance.
Publication collections
Topics
- Adversarial Robustness and Privacy
- AI for Asset Management
- AI for Business Automation
- AI for Code
- AI for Supply Chain
- AI Testing
- Automated AI
- Causality
- Computer Vision
- Conversational AI
- Explainable AI
- Fairness, Accountability, Transparency
- Foundation Models
- Human-Centered AI
- Knowledge and Reasoning
- Machine Learning
- Natural Language Processing
- Neuro-symbolic AI
- Speech
- Trustworthy AI
- Trustworthy Generation
- Uncertainty Quantification
