As B Corp organisations embrace artificial intelligence, they face a difficult question: What does responsible AI use actually look like when your values demand you consider more than the bottom line?
Several B Corp organisations recently highlighted how they navigate AI use to ensure accountability, measurability, and ethical reasoning. Implementing AI in business functions is complex; however, some organisations are managing to balance the pros and cons.
For Beneficial State Bank, the answer isn't simple.
The Oakland, California-based community development bank is a certified B Corporation built around social justice, climate accountability, and financial success, without financing fossil fuels or private prisons. AI could help advance those goals by making lending more efficient, identifying hidden bias, and extending capital to communities that have historically been shut out of the financial system.
But using AI also comes with environmental and social costs; for a company whose purpose is to create a more equitable and sustainable economy, that creates a genuine dilemma.
Until recently, Terra Neilson, Beneficial State Bank's chief impact officer, said it was easy to put AI in the "unethical" category. The technology's environmental footprint, energy demands, and potential social harms have become increasingly difficult to ignore. Then came a project that complicated that conclusion.
Beneficial State Bank recently extended a pilot launched by Beneficial State Foundation, its non-profit majority shareholder, using Stratyfy's AI-assisted credit decisioning technology to improve efficiency and identify systemic bias in credit underwriting.
The potential impact is significant.
BetterFi, a Tennessee-based community development financial institution participating in the pilot, has already used the technology to increase approvals among BIPOC communities by 21%.
For Neilson, that raises an important possibility: What if AI can help financial institutions see and correct the blind spots that have contributed to redlining and unequal access to capital for generations?
"If the banking system can use AI to find and correct those blind spots," the thinking goes, perhaps the technology can become part of the solution rather than simply another source of harm.
But that doesn't make the decision easy.
"You get decision fatigue," Neilson says. "We're being really intentional in this moment to say, what are the harms and what are the benefits that are worth the costs that we see?"
That may be the defining question for mission-driven businesses entering the AI era.
AI Adoption for a B Corp
For a B Corp, adopting AI isn't just a productivity decision; it's a stakeholder decision. Every tool carries environmental, social, and workforce impacts.
"AI is definitely in a world of its own because of the potential to bury the trail between decision and impact," Neilson says.
That matters when AI influences lending, hiring, marketing, or strategy. For organisations built on transparency and accountability, understanding those impacts is essential.
B Lab's updated standards treat AI as a stakeholder governance issue: companies should consider where their data comes from, how models are hosted, their environmental footprint, and who benefits or bears the costs.
"We don't want to say you can or can't use it," says B Lab's Clay Brown. "We want to say you need to understand the impacts ... and take steps to mitigate them."
Measurability Issues
The challenge is that companies can't manage what they can't measure.
Alex de Vries-Gao, a researcher at VU Amsterdam, says AI suppliers still provide limited information about their environmental impacts. That leaves mission-driven companies balancing the benefits of AI against costs they may not be able to quantify.
Do they use AI and accept the uncertainty, or avoid it and risk falling behind?
Responsible AI doesn't have to mean no AI.
For some B Corps, the answer is more deliberate use rather than a blanket rejection of AI.
London-based B Corp, Third City, evaluates AI providers against its values while acknowledging the technology's environmental cost. Its GEOView tool, for example, requires repeated searches across AI platforms, increasing its footprint.
The firm uses Ecolytics' Offset AI to track estimated carbon and water impacts and purchase environmental offsets.
"Obviously, offsetting has a mixed reputation amongst the B community," says founding partner Mark Lowe. "But you've got to do something."
For mission-driven companies, waiting for perfect information may be less responsible than acting on what they know today.
Using AI thoughtfully
Yulu, a Vancouver-based B Corp, is taking a different approach: using less AI and using it more thoughtfully.
Employees are encouraged to refine complex prompts with colleagues before turning to AI, reducing unnecessary back-and-forth and compute.
The company has also embedded its ethical AI principles directly into its internal AI tools, making responsible use the default rather than another policy employees must remember.
Yulu updates its guidelines every six months as the technology and its understanding of the risks evolves.