What is artificial intelligence explainability?
A working definition from MIT Sloan
artificial intelligence explainability (noun)
A quality that enables users of artificial intelligence programs to understand and trust how models operate and make decisions.
Creating successful artificial intelligence programs doesn’t end with building the right system. Stakeholders must have confidence that the programs are accurate and trustworthy.
According to research from Ida Someh, Barbara Wixom, and Cynthia Beath of the MIT Sloan Center for Information Systems Research, artificial intelligence explainability helps by ensuring users that models are “value-generating, compliant, representative, and reliable.”
There are several reasons stakeholders hesitate to trust AI. Because AI is relatively new, there isn’t an extensive list of proven use cases. Models are often opaque — AI relies on complex math and statistics, so it can be hard for average users to tell how a model works, whether it is producing accurate results, and if it is ethical and compliant.
Models can produce biased results if trained on biased data, and they also “drift” over time, meaning they can start producing inaccurate results as the world changes or incorrect data is included and replicated.
AI explainability is an emerging field, and teams working on AI projects are mostly creating the playbook as they go, the researchers write. Organizations can start by identifying units and organizations that are already creating effective AI explanations, continuing to test the most promising practices, and institutionalizing the best ones.
Working Definitions: Artificial Intelligence
MIT Sloan's Working Definitions explore the words and phrases behind emerging management ideas.
Leading the AI-Driven Organization
In person at MIT Sloan
Register Now
‘No One Works Here’ urges firms to tackle human bottlenecks with AI
In his new book, MIT Sloan School of Management’s Paul Cheek argues that companies lose competitive ground when they keep humans in the loop on decisions that AI can make more efficiently.
4 ways to navigate entrepreneurship in the age of AI
MIT Sloan’s Bill Aulet and Jenny Larios Berlin explore how founders can combine artificial intelligence with the principles of Disciplined Entrepreneurship to move from concept to market more quickly.
An MIT expert on which companies will succeed in the AI era
MIT Sloan’s Andrew McAfee said the future of tech-driven companies is young firms on the West Coast — and that Europe will continue to fall behind.
AI ethics and governance: Where will you draw the line?
Enterprises must address complex ethical issues surrounding AI or risk exposure to myriad financial, legal, and reputational risks. Here’s a framework to protect your organization.
Can AI’s climate benefits outweigh its costs?
The rising energy demands of AI and data centers threaten corporate net zero commitments and worsen climate change. But can AI also help develop climate solutions? MIT Sloan researchers investigate.
Three keys to building a semantic layer
A new research briefing explains how semantic layers make enterprise data more accessible and understandable by humans and machines.
How will AI automation hit — like a crashing wave or a rising tide?
AI performance is improving across many workplace tasks broadly, not in sudden shocks, new MIT research finds. That cadence gives companies and workers time to prepare for task-level displacements.
6 questions to guide your AI strategy
Few AI pilots scale successfully. Asking key questions about goals, governance, and culture can help them thrive.
Why AI-driven enterprises are the future of entrepreneurship
AIDEs have lower overhead, are extremely efficient at product development, require a smaller footprint, and provide increased ownership for founders, according to MIT Sloan’s Paul Cheek.
How organizations can capture value from digital colleagues
As AI systems become more capable, companies will need to think beyond individual tools and begin managing AI as a digital colleague — part of how teams, workflows, and business processes operate.