AI is not the biggest risk to mortgage lending right now. That is a policy misunderstanding.
While lenders across the country are moving quickly to implement artificial intelligence responsibly, state lawmakers are moving just as quickly to regulate it. And when policy moves faster than policymakers’ understanding, unintended consequences follow.
As former Cisco CEO John Chambers recently noted: “AI moves five times faster and will deliver three times as many results as the Internet age.” The mortgage industry is already feeling that acceleration in real time. Artificial intelligence is rapidly changing the way lenders detect fraud, ensure compliance, improve accuracy and provide borrowers with faster decisions.
But while technology is rapidly evolving, policy is trying to keep pace – and not always in a way that supports consumers or lenders.
AI is already embedded in modern mortgage lending
For mortgage professionals, artificial intelligence is not theoretical or futuristic. It is already embedded in the entire loan lifecycle.
Automated underwriting systems, fraud detection tools, service analytics and customer engagement platforms all rely on advanced data models and automated decision making. These tools help lenders operate securely, consistently and at scale while meeting increasingly complex regulatory expectations.
They also support the very outcomes policymakers want to protect:
- More accurate credit decisions
- Stronger supervision of fair lending
- Faster approvals
- Fewer operational errors
- Fraud prevention
- Removing human prejudices
- Better consumer experiences
When AI is used responsibly, it improves compliance rather than undermines it. It allows lenders to monitor results more accurately, identify risks earlier and ensure greater transparency throughout the process.
But as these technologies evolve, lawmakers are increasingly focused on how to govern — and in some cases restrict — them.
The California policy moment: AB 1018
Last year, California lawmakers considered AB 1018, a proposal aimed at preventing “algorithmic discrimination” in automated decision-making systems.
The goal was valid and widely supported. Protecting consumers from bias and ensuring fair outcomes must remain central to any conversation about AI in mortgage lending.
But as initially drafted, AB 1018 created significant operational concerns for financial institutions that were already operating under extensive federal supervision. The bill proposed new state-level requirements on top of existing federal laws related to privacy, data protection, fair lending, and model management.
In practice, dual regulatory frameworks create complexity rather than clarity.
Mortgage lenders already operate in one of the most regulated sectors of the economy. Federal requirements set clear expectations around data use, consumer protection and model management. Adding additional state-specific rules on top of these frameworks could introduce conflicting standards, increase compliance costs, and delay the implementation of technologies that improve both efficiency and consumer outcomes.
Most importantly, this risks limiting access to credit, especially for borrowers in underserved communities.
When policy exceeds operational reality
If policies inadvertently restrict or complicate the use of these tools, the impact extends far beyond lenders.
Approvals take longer.
Operational costs increase.
Investments in technology are slowing down.
And ultimately, borrowers face reduced access to efficient, affordable credit.
This is not a theoretical concern. It is the operational reality that lenders face when well-intentioned legislation is developed without sufficient involvement from those responsible for its implementation.
The conversation surrounding AB 1018 made one thing clear: policymakers and industry leaders must work together sooner and more to ensure outcomes that truly benefit consumers.
Engagement changes outcomes
There is also an encouraging lesson in the way this issue has developed.
The involvement of the sector made a difference.
Organizations like the California MBA worked directly with legislators and legislative staff to provide operational context, clarify existing regulatory frameworks, and propose constructive adjustments. Those conversations helped narrow the proposal and align it with existing federal oversight.
This kind of engagement is not about resisting regulation. It’s about improving it.
Thoughtful, effective policies emerge when lawmakers understand how mortgage lending actually works: how decisions are made, how data is managed, and how compliance obligations intersect with technology implementation. Achieving this level of understanding only happens when industry participants have a seat at the table.
AB 1018 could be revisited in the 2026 legislative session, through amendments or new legislative proposals addressing AI and automated decision-making systems. Similar conversations are already emerging in other states and at the federal level.
As California moves, the nation watches
California has long served as a policy indicator. Regulatory frameworks that start here often influence national and even international approaches.
That reality creates both risks and opportunities.
Artificial intelligence provides enormous efficiency and economic growth, but also raises legitimate questions around consumer protection, transparency and data security. Policymakers are eager to act – and the frameworks they build will shape the mortgage industry for years to come.
If regulatory structures support responsible innovation, lenders will be able to continue to improve accuracy, expand access to credit and deliver evolved consumer experiences. If these structures create unnecessary duplication or limit the use of proven technologies, the industry will experience higher costs and reduced flexibility – outcomes that will ultimately impact borrowers.
This isn’t just a California issue. It is a national business issue for lenders, service providers and technology providers alike.
Advocacy is now a business strategy
The mortgage industry has always adapted to changes. But in the current environment, adaptation alone is not enough.
Artificial intelligence will continue to transform the way lenders work. At the same time, policymakers will continue to work to understand and regulate its use.
In this environment, passive observation is not a viable strategy.
Involvement must be proactive.
Conversations need to start earlier.
Leaders must recognize that advocacy is no longer separate from business strategy – it is an essential part of it.
At the California MBA, we want to ensure that AI policy discussions remain grounded in operational realities, consumer outcomes, and the need for a modern, efficient mortgage system.
The future of mortgage lending will not be determined by technology alone; it will be shaped by the policy decisions being written right now – and by the leaders willing to help shape them.
Paul Gigliotti is the CEO of California MBA, where he leads one of the nation’s most influential state mortgage associations at the intersection of advocacy, policy and industry strategy.
This column does not necessarily reflect the opinion of HousingWire’s editorial staff and its owners. To contact the editor responsible for this piece: [email protected].
Related
#policy #practice #mortgage #industry