Over the last week, millions of people around the world have interacted with OpenAI’s ChatGPT, which represents a significant advance for generative artificial intelligence (AI) and the foundation models that underpin many of these use cases. It’s a fitting way to end what has been another big year for the industry.
We’re at an exciting inflection point for AI. Adoption of AI among businesses is increasing, and research and AI development on foundation models is enabling use cases like generative AI to become even more sophisticated and powerful. The potential is vast. It can help us leverage significant amounts of data to start designing and discovering new solutions to business and societal problems such as those related to sustainability, life sciences, customer care, employee experience and many more.
These advances are simultaneously raising separate, but important discussions and questions within the industry: how can you trust the algorithms and outputs of these models? How can we ensure that these models are being used responsibly? For example, generative AI models can produce highly believable, well-structured responses so it can be hard to immediately pinpoint an incorrect response without the right subject matter expertise.
This is a dialogue that IBM is engaging in with our clients and partners every day. Advances across AI technology are happening quickly. At the same time, governments around the world are continuously evaluating and implementing new AI guidelines and AI regulation frameworks. We think that businesses have an opportunity to act now to put guardrails in place internally to govern how AI is developed and deployed.
To scale the use of responsible AI requires AI governance, the process of defining policies and establishing accountability throughout the AI lifecycle. This can also help your models adhere to principles of fairness, explainability, robustness, transparency and privacy. A comprehensive AI governance strategy encompasses people, process and technology.
AI governance technology can help implement guardrails at each stage of the AI/ML lifecycle. This includes data collection, instrumenting processes and transparent reporting to make needed information available for stakeholders. We recently launched IBM AI Governance, a solution designed to help companies get a better understanding of what’s going on below the surface of these systems. IBM AI Governance is designed to help businesses develop a consistent transparent model management process, capturing model development time, metadata, post-deployment model monitoring and customized workflows. IBM has also developed and open sourced a set of Trusted AI toolkits, including AI Fairness 360, Adversarial Robustness 360, AI Explainability 360, Uncertainty Quantification 360 and AI FactSheets 360.
In addition to discussions on the importance of AI governance, many of the advances across the industry reinforce IBM’s focus on the unique needs of AI for business. Our clients want AI that is designed for and managed by their subject matter experts, and that can be easily customized based on their domain and business priorities; that is robust with high accuracy and reliability; that operates in and navigates through siloed data in complex formats; and that is guided by principles of trust and transparency. This focus on AI for business is what guides the development of our AI software like IBM Watson Assistant and IBM Watson Discovery.
2022 has been another big year for AI with increasing adoption across the industry as well as promising new advancements. We believe that businesses that embrace AI governance early will be better positioned to responsibly harness this technology now and in the future.
Source: ibm.com
Posts
0 comments:
Post a Comment