Tools & Platforms
UCR pioneers way to remove private data from AI models | UCR News
A team of computer scientists at UC Riverside has developed a method to erase private and copyrighted data from artificial intelligence models—without needing access to the original training data.
This advance, detailed in a paper presented in July at the International Conference on Machine Learning in Vancouver, Canada, addresses a rising global concern about personal and copyrighted materials remaining in AI models indefinitely—and thus accessible to model users—despite efforts by the original creators to delete or guard their information with paywalls and passwords.
The UCR innovation compels AI models to “forget” selected information while maintaining the models’ functionality with the remaining data. It’s a significant advancement that can amend models without having to re-make them with the voluminous original training data, which is costly and energy-intensive. The approach also enables the removal of private information from AI models even when the original training data is no longer available.
“In real-world situations, you can’t always go back and get the original data,” said Ümit Yiğit Başaran, a UCR electrical and computer engineering doctoral student and lead author of the study. “We’ve created a certified framework that works even when that data is no longer available.”
The need is pressing. Tech companies face new privacy laws, such as the European Union’s General Data Protection Regulation and California’s Consumer Privacy Act, which govern the security of personal data embedded in large-scale machine learning systems.
Moreover, The New York Times is suing OpenAI and Microsoft over the use of its many copyrighted articles to train Generative Pre-trained Transformer, or GPT, models.
AI models “learn” the patterns of words from a vast amount of texts scraped from the Internet. When queried, the models predict the most likely word combinations, generating natural-language responses to user prompts. Sometimes they generate near-verbatim reproductions of the training texts, allowing users to bypass the paywalls of the content creators.
The UC Riverside research team—comprising Başaran, professor Amit Roy-Chowdhury, and assistant professor Başak Güler—developed what they call a “source-free certified unlearning” method. The technique allows AI developers to remove targeted data by using a substitute, or “surrogate,” dataset that statistically resembles the original data.
The system adjusts model parameters and adds carefully calibrated random noise to ensure the targeted information is erased and cannot be reconstructed.
Their framework builds on a concept in AI optimization that efficiently approximates how a model would change if it had been retrained from scratch. The UCR team enhanced this approach with a new noise-calibration mechanism that compensates for discrepancies between the original and surrogate datasets.
The researchers validated their method using both synthetic and real-world datasets and found it provided privacy guarantees close to those achieved with full retraining—yet required far less computing power.
The current work applies to simpler models—still widely used—but could eventually scale to complex systems like ChatGPT, said Roy-Chowdhury, the co-director of UCR’s Riverside Artificial Intelligence Research and Education (RAISE) Institute and a professor in the Marlan and Rosemary Bourns College of Engineering.
Beyond regulatory compliance, the technique holds promise for media organizations, medical institutions, and others handling sensitive data embedded in AI models, the researchers said. It could also empower people to demand the removal of personal or copyrighted content from AI systems.
“People deserve to know their data can be erased from machine learning models—not just in theory, but in provable, practical ways,” Güler said.
The team’s next steps involve refining the method to work with more complex model types and datasets and building tools to make the technology accessible to AI developers worldwide.
The title of the paper is “A Certified Unlearning Approach without Access to Source Data.” It was done in collaboration with Sk Miraj Ahmed, a computational science research associate at the Brookhaven National Laboratory in Upton, NY, who received his doctoral degree at UCR. Both Roy-Chowdary and Güler are faculty members in the Department of Electrical and Computer Engineering with secondary appointments in the Department of Computer Science and Engineering.
San Jose, California, in the heart of Silicon Valley, is testing new artificial intelligence software designed to speed up the city’s building permit process, with the goal of making it easier to build residential housing.
The pilot program, launched Thursday by the city’s Planning, Building and Code Enforcement Department, uses AI from CivCheck to pre-check applications and flag problems before they’re submitted. According to the department’s Housing Production Dashboard, more than 90% of accessory dwelling unit applications are returned to applicants due to missing information, which can delay construction. The city aims to build 62,200 units by 2031.
The pilot is part of San Jose’s larger push to use technology across government, from optimizing bus routes to installing license plate readers.
Tasha Dean, chief communications officer for Mayor Matt Mahan, called the effort a step toward “bringing government into the 21st century.”
“I think that most residents interact with their government in a few ways, right? They wait in line at the DMV, they come to their city council meeting and sit through hours of back and forth with elected officials, or they apply for and they spend weeks going back and forth with planners. These are all things that are incredibly frustrating — they’re slow, and ultimately, they reduce trust in government,” Dean said in an interview, “We’re trying to change that view [and] prove that government can be responsive to the needs of the people.”
The testing phase for building permits will begin this fall, with staff using the tool alongside manual reviews. If the pilot project proves successful, Dean said, the city will release it to the public as early as next year, first for accessory dwelling unit permits, which account for roughly a quarter of new-build permit applications, and eventually for single-family homes.
“I mean, part of the big delays and the issues that makes the housing crisis so bad in California is how long it takes,” she said. “And if we can reduce the friction there, ultimately we hope that we can get more shovels in the ground.”
Dean said the AI system could also help the city move quickly in emergencies, similar to the tool California Gov. Gavin Newsom announced last May, designed to expedite the building permit approval process in Los Angeles after the devastating Eaton and Palisades wildfires in January, which destroyed more than 16,000 homes, businesses and other buildings.
Balancing growth against risk is nothing new to finance chiefs, but as CFOs look to navigate continuing economic uncertainty, alongside regulatory change, it’s a costly challenge that’s only grown more complex. The size of the governance, risk and compliance solutions market jumped over 14% to reach a valuation of $51.5 billion as of early 2025, as companies seek to hedge against regulatory and economic shifts as well as technology changes and risks, according to an April article by Securify.
In this environment, many finance chiefs are also being asked to cut down on costs amid ongoing pricing pressures, while at the same time, to find the space and resources needed to integrate emerging technologies such as artificial intelligence into their businesses.
That puts many finance chiefs in a bind: while “every CFO wants to leverage the savings associated with enhanced productivity through AI,” they also have to account for the risks that come alongside the technology — which are “enormous,” Raul Villar Jr., CEO of audit and compliance platform AuditBoard told CFO Dive.
Such risks include everything from false positives or “hallucinations” generated by AI tools, to fraud and cyberattacks utilizing the technology to unsafe practices by employees using the technology. As such, the “number one component of risk that CFOs are looking at today” surrounds where AI sits inside of their organization, Villar Jr. said in an interview.
Unraveling the AI risk tangle
Business and technology leaders have continued to shine a spotlight on AI’s potential, with many looking to tap the technology to help solve some of the challenges faced by continued economic headwinds. AI spending, especially on solutions like “agentic AI” tools that can perform tasks with little or no human supervision, has continued to jump in recent months, with a growing number of companies investing $10 million or more in the tool, CFO Dive previously reported.
In the face of all that attention, there’s “so much pressure for every CFO to be leveraging AI,” Villar Jr. said. If you’re a CFO at a public company, “it comes up on every public call. If you’re [in] a private company, your board is definitely asking you. It’s top of mind,” he said.
Villar Jr. has served as CEO for the Los Angeles, California-based AuditBoard — which offers audit and compliance software — since July of this year, according to his LinkedIn profile. Prior to the platform, he served as executive chair for business management software provider Simpro. Past roles also include a 21-year span at human resource software company APD, where he held a number of positions including serving as its SVP, sales, major account services.
As finance chiefs mull the potential return on investment of AI, however, the risks — and potential costs —associated with the technology are looming large in their minds as they seek to navigate a host of economic and regulatory challenges, including potential shifts to how AI itself is regulated.
That means CFOs have to strategically “pick their spots” when it comes to investing and integrating AI tools into their organizations, Villar Jr. said. To do so effectively, they don’t need to just understand AI capabilities and their associated risks, but the data that is being fed into the technology and the risks associated with those data sources, he said.
Keeping the focus on data
As AI becomes more commonplace inside of businesses, CFOs will face a twofold challenge: they will need to ensure the security of the data being fed into emerging AI tools, while also being able to quickly parse the insights they need from that data, he said.
“Protecting your data is critical, but turning data into insights is probably the most important thing a CFO has to learn to do because we’ve all gotten accustomed to a certain data layer…that [companies] report on,” he said.
As the technology matures, CFOs should approach their AI investment decisions similarly to how they approach headcount decisions: “It’s just another lever to pull from a productivity perspective, but each function is different and you’re going to have to pick winners and losers within your firm and test and manage quickly,” Villar Jr. said.
Tools & Platforms
Trotter Overhead Door Becomes First Oklahoma Company to Implement AI-Powered Diagnostic Technology for Precision Garage Door Opener Installation Services
Company vehicle and branding from Trotter Overhead Door Garage + Home, representing their trusted garage and home solutions.
Trotter Overhead Door Garage + Home, serving Oklahoma City since 1983, is the first in the state to use AI diagnostic technology for garage doors. The new system ensures precise, reliable opener installations by analyzing door specs, weight, and usage patterns.
Richardson, TX – Trotter Overhead Door Garage + Home has announced the implementation of groundbreaking artificial intelligence diagnostic technology, positioning the company as Oklahoma’s first garage door service provider to integrate AI-powered precision systems into daily operations. This technological advancement represents a significant milestone for the family-owned business that has served the Oklahoma City metropolitan area since 1983.
The innovative AI system analyzes door specifications, weight distribution, and operational patterns to ensure optimal performance and longevity. This technology particularly benefits garage door opener installation projects by providing precise compatibility assessments and optimal configuration settings. This development comes as property owners increasingly seek reliable solutions for garage door repair services that minimize future maintenance requirements.
Revolutionary Technology Transforms Service Delivery
The new AI diagnostic platform utilizes advanced sensors and machine learning algorithms to evaluate garage door systems before service begins. This technology provides detailed assessments of mechanical components, electrical systems, and structural integrity, allowing technicians to identify potential issues before they become costly problems.
Property owners throughout the Oklahoma City area can now benefit from enhanced accuracy in service recommendations and installation procedures. The system’s diagnostic capabilities extend beyond standard visual inspections, offering comprehensive analysis that improves both safety and performance outcomes for overhead door installations in OKC.
Enhanced Safety Protocols Through Smart Technology
Safety remains paramount in all garage door operations, and the AI system introduces multiple layers of protection for both technicians and property owners. The technology continuously monitors installation processes, alerting technicians to potential safety hazards or installation irregularities in real-time.
The diagnostic system evaluates spring tension, cable integrity, and opener compatibility to prevent accidents and equipment failures. This proactive approach has already shown promising results in reducing service callbacks and improving customer satisfaction rates across garage door repair projects in OKC.
Family-Owned Innovation Meets Modern Technology
Despite embracing modern technology, Trotter Overhead Door Garage + Home maintains its commitment to family values and personalized service. Jesse and Tina Trotter established the company with a vision of combining superior products with exceptional customer care, principles that continue to guide operations today.
The integration of AI technology enhances rather than replaces the human expertise that has made the company successful. Skilled technicians now have access to precise diagnostic information that enhances their ability to recommend appropriate solutions for each unique situation requiring overhead door services in Oklahoma City.
Comprehensive Service Platform Expansion
The AI diagnostic system supports the company’s full range of services, from routine maintenance to complex installations. Property owners can schedule comprehensive evaluations that provide detailed reports on the health of their garage door system, including recommended improvements and preventive maintenance schedules.
This technology particularly benefits commercial properties that require reliable access and security. The system’s ability to predict maintenance needs helps business owners avoid unexpected downtime and costly emergency repairs, making it a valuable tool for planning garage door repair in OKC.
Training and Certification Programs Launch
To support the new technology implementation, Trotter Overhead Door Garage + Home has developed comprehensive training programs for all service technicians. These programs ensure consistent application of AI diagnostic procedures across all service calls and installations.
The training curriculum includes both the technical operation of the diagnostic equipment and the interpretation of data analytics. This investment in staff development reinforces the company’s commitment to maintaining service quality standards while adapting to technological innovations in overhead door solutions.
Future Impact on the Oklahoma Garage Door Industry
The implementation of AI diagnostic technology positions Trotter Overhead Door Garage + Home at the forefront of industry innovation in Oklahoma. This advancement demonstrates how family-owned businesses can successfully integrate modern technology while preserving traditional values of quality and service.
The company’s 15,000-square-foot facility at 14000 North Santa Fe Avenue in Edmond now serves as a technology hub for garage door services throughout the region. Property owners can visit the showroom to experience the diagnostic technology firsthand and learn about available service options.
Industry analysts expect this technological advancement to influence service standards throughout the Oklahoma garage door market, potentially inspiring similar innovations among other service providers.
About Trotter Overhead Door Garage + Home
Established in 1983 by Jesse and Tina Trotter, Trotter Overhead Door Garage + Home has provided garage door solutions to the Oklahoma City metropolitan area for over four decades. The family-owned company operates from a 15,000-square-foot facility in Edmond, offering installation, repair, and maintenance services for both residential and commercial properties. The company maintains core values of integrity, quality, service, continuous learning, care, and excellence in all operations.
For more information about AI diagnostic services or to schedule a consultation, please get in touch with Trotter Overhead Door Garage + Home at (405) 341-9601 or visit our showroom located at 14000 North Santa Fe Avenue, Edmond, Oklahoma.
Media Contact
Company Name: Trotter Overhead Door Garage + Home
Contact Person: Nate
Email: Send Email
Phone: (405) 341-9601
Address:14000 North Santa Fe Ave.
City: Edmond
State: Oklahoma 73013
Country: United States
Website: https://www.trotteroverheaddoor.com/
-
Business6 days agoThe Guardian view on Trump and the Fed: independence is no substitute for accountability | Editorial
-
Tools & Platforms3 weeks ago
Building Trust in Military AI Starts with Opening the Black Box – War on the Rocks
-
Ethics & Policy1 month agoSDAIA Supports Saudi Arabia’s Leadership in Shaping Global AI Ethics, Policy, and Research – وكالة الأنباء السعودية
-
Events & Conferences4 months agoJourney to 1000 models: Scaling Instagram’s recommendation system
-
Jobs & Careers2 months agoMumbai-based Perplexity Alternative Has 60k+ Users Without Funding
-
Education2 months agoVEX Robotics launches AI-powered classroom robotics system
-
Funding & Business2 months agoKayak and Expedia race to build AI travel agents that turn social posts into itineraries
-
Podcasts & Talks2 months agoHappy 4th of July! 🎆 Made with Veo 3 in Gemini
-
Podcasts & Talks2 months agoOpenAI 🤝 @teamganassi
-
Education2 months agoMacron says UK and France have duty to tackle illegal migration ‘with humanity, solidarity and firmness’ – UK politics live | Politics