Artificial Intelligence

George �鶹��Ƶ launches its inaugural ACCESS Academy cohort

Katarina Benson — Wed, 09 Jul 2025 20:09:41 +0000

George �鶹��Ƶ launches its inaugural ACCESS Academy cohort Katarina Benson Wed, 07/09/2025 - 16:09

The College of Education and Human Development (CEHD) at �鶹��Ƶ is excited to welcome the inaugural class of students for their Accelerated College and Employability Skills (ACCESS) Academy.

Developed in partnership with Northern Virginia Community College (NOVA) and Loudoun County Public Schools (LCPS), ACCESS Academy is a lab school designed to reimagine how students prepare for college, careers, and civic life in a rapidly evolving world. The program will particularly support students who may not have previously expected to follow a college pathway—those navigating limited exposure to higher education or career pathways, relevant information, or professional networks.

“ACCESS Academy reflects a commitment to transforming how we think about readiness—not just for college or careers, but for life in a complex, interconnected society,” said Ingrid Guerra-López, dean of CEHD. “By aligning K–12, higher education, and industry partners, we’re creating more coherent and responsive pathways that honor learners’ potential while addressing workforce and societal needs.”

ACCESS Academy students will work toward high school graduation requirements while engaging in real-world problem-solving situations informed by industry partners. They will develop future-ready skills, including critical thinking, collaboration, and adaptability, while building the mindsets and capacities required to navigate dynamic and evolving work environments.

"With the launch of ACCESS Academy, we’re not just starting a new program, we’re opening doors to new possibilities for dual enrolled students and reimagining what learning could look like. This inaugural class is stepping into a bold, future-focused initiative designed to empower students not only to succeed in high school, but to lead in college, career and beyond,” said Eun-Woo Chang, NOVA vice president of academic affairs and chief academic officer.

As a preview of ACCESS Academy’s learning model, students from across Northern Virginia are participating in the ACCESS Academy Artificial Intelligence (AI) Summer Camp at �鶹��Ƶ Square in Arlington. This one-week program introduces foundational concepts in machine learning through engaging hands-on projects guided by problem-based learning strategies.

“The level of interest in the ACCESS Academy AI Summer Camp has been both affirming and encouraging,” said Roberto Pamas, director of ACCESS Academy. “With three week-long sessions accommodating a total of 180 students, the camp presents an early

opportunity for students to engage with foundational concepts in machine learning while offering a preview of ACCESS Academy’s broader approach to experiential education.”

ACCESS Academy’s curriculum is shaped by collaboration among faculty in CEHD and the Department of Information Sciences and Technology at the College of Engineering and Computing, with input from NOVA, LCPS, and industry partners who make up the ACCESS Academy Curriculum Innovation Advisory Board. This collaborative approach is designed to ensure alignment with evolving workforce needs while streamlining coursework.

“The insights shared through our collaboration with the advisory board have been invaluable,” said Guerra-López. “They help ensure that ACCESS Academy remains grounded in what will best prepare our graduates to be competitive and confident in the modern workforce, even if that requires reimagining traditional educational pathways and practices.”

“ACCESS Academy is more than a program—it’s a launchpad. By combining academic rigor with real-world relevance, we’re empowering students to graduate with the skills, experiences, and credentials needed to thrive in tomorrow’s workforce,” said LCPS Chief Academic Officer Ashley Ellis.

ACCESS Academy is positioned for expansion through the formation of a regional consortium that could include new partnerships with school divisions across Virginia’s Region 4 to scale the model for other high-demand fields such as education and health care.

Powering Artificial Intelligence at George �鶹��Ƶ

Learn more about ACCESS Academy

Topics

Campus News

College of Education and Human Development

K-12 Partnerships

Fuse at �鶹��Ƶ Square

Artificial Intelligence

AI4Defense develops DoD-driven smart solutions

Nathan Kahl — Fri, 23 May 2025 11:20:18 +0000

AI4Defense develops DoD-driven smart solutions Nathan Kahl Fri, 05/23/2025 - 07:20

Kammy Sanghera, executive director of global engagements and tech talent development and a professor in the Department of Information Sciences and Technology at �鶹��Ƶ, received a grant from the Griffiss Institute to lead the AI4Defense program—an innovative incubator that empowers high school and early undergraduate students to explore artificial intelligence (AI) applications in national defense.

Funded by the Air Force Office of Scientific Research, the grant supports AI4Defense, a month-long, hands-on experience designed to immerse students in real-world problem-solving. Participants collaborate with mission partners—including the Department of Defense (DoD) and agencies such as the Space Force and the Defense Digital Service—to develop AI-driven solutions to national security challenges.

The Griffiss Institute, a leading STEM accelerator for the DoD and the Defense Innovation Unit (DIU), help expand the program's reach and impact. “A very important element happened last year,” said Sanghera. “DIU joined forces with the same mindset—build the talent pipeline, engage mission partners to solve real-world problems, and use AI to create efficient and secure solutions. These three elements were perfect for the design of the program.”'

Students at the 2024 AI4Defense event. Photo provided.

With support from academic and industry mentors, the program has grown into a public–private partnership that Sanghera and her team aim to scale further this summer. Students not only gain technical and communication skills but also engage in customer discovery—working directly with mission partners to define problems and tailor AI solutions accordingly.

In 2024, one student team developed a proof-of-concept AI agent for the U.S. Air Force to automate the creation of Self-Assessment Checklists (SACs). Traditionally a manual and time-intensive process due to regulatory complexity and frequent updates, the AI solution was able to extract key phrases from regulations and generate relevant questions, significantly improving the speed, accuracy, and consistency of SAC creation.

Sanghera emphasized the program’s multilevel impact: George �鶹��Ƶ undergraduate and graduate students served as mentors to high school participants, gaining leadership experience and exposure to the defense innovation ecosystem.

“Many of our graduate students are international students, so it was also a valuable opportunity to work alongside early undergraduate students and with unclassified data and explore career pathways in the defense sector,” Sanghera said.

The Defense Department strategy for AI adoption. Photo credit Department of Defense.

The program was promoted across Fairfax County, Arlington County, and Loudoun County public schools and culminated in a final symposium. Students pitched their AI solutions to a panel of judges, received feedback, and were recognized with awards. Several teams chose to continue collaborating with their mission partners beyond the summer program.

This year's program began in June, and Sanghera is enthusiastic about the possibilities. “We’re excited to see what creative, impactful solutions students bring forward.”

Associate Professor Mihai Boicu—Co-PI with Sanghera on the AI4Defense initiative—lead efforts to guide students in presenting their work through a continuing research lens. He also oversees the development of peer-reviewed publications based on students’ projects, which will be cataloged as part of the program’s scholarly output and shared broadly to promote AI learning and innovation in future cohorts.

Sanghera and Boicu will continue collaborating with Caitlin Bowers, a Defense Innovation Faculty Fellow at George �鶹��Ƶ. Together, Sanghera’s strategic vision, Boicu’s research expertise, and Bowers’ experience in defense innovation projects form a strong team that will provide students with the opportunity to gain practical skills and apply them to real-world challenges relevant to the defense sector.

In This Story

Kamaljeet Sanghera

Mihai Boicu

Learn more about Research at George �鶹��Ƶ

Topics

U.S. Department of Defense

Research

Artificial Intelligence

K-12 Partnerships

Defense and Security

Trial by virtual fire: Augmented reality project tailors training for firefighters

Sarah Holland — Thu, 22 May 2025 17:18:32 +0000

Trial by virtual fire: Augmented reality project tailors training for firefighters Sarah Holland Thu, 05/22/2025 - 13:18

For firefighters, even training is a risk. According to the United States Fire Administration, 16.6% of all firefighter fatalities in 2024 took place during training exercises. �鶹��Ƶ researcher Craig Yu is hoping to change that statistic.

Yu, associate professor of computer science, has been developing augmented and virtual reality training for most of his career, and he has a particular interest in exploring how artificial intelligence (AI) and emerging technologies could help research on human development and sports science. Through National Science Foundation’s EArly-concept Grants for Exploratory Research (EAGER) program, Yu and Joel Martin, associate professor of kinesiology, are developing just that: an augmented reality program that could help train firefighters while minimizing risk.

“Firefighters have high rates of injuries and many injuries happen during training. There is a need for training technology that can replicate aspects of emergencies while minimizing the injury risk and physical burn,” said Martin. Martin has spent the past few years working with Fairfax County Fire and Rescue to determine how they can reduce physical strain to firefighters during training exercises. “I think that technology such as immersive reality training can help to meet training needs while minimizing the physical risks of performing certain drills."

A prototype of the virtual training space from their proposal. Image provided.

Augmented—or mixed—reality uses devices such as phones or a headset to overlay 3D computer-generated content onto the real world. An increasingly accessible tool, you can find augmented reality in applications like Ikea furniture previews in your house, dinosaur encounters in the Smithsonian, and catching Pokémon in your neighborhood.

As described in their project proposal, “Adapting Mixed Reality Training Programs to Real-World Scenes to Enhance Human-AI Teaming in Emergency Responses,” the new training would utilize augmented reality to create realistic training scenarios for firefighters.

Based on a 3D digital twin of the room, the program will use AI to analyze what's in the room and create a realistic outbreak of fire: where it is most likely to start and spread, and where civilians are most likely to be.

“The program understands the environment, and it will be refined with additional input from professional firefighters,” said Yu.

Yu also notes that the scenario can be adapted for different experience levels. Simpler scenarios can be developed for new firefighters, which can then be adjusted for increasing sophistication and complexity as the trainees learn and improve. It also allows them to practice scenarios that would be difficult or impossible to replicate in a live burn, such as how to best evacuate and rescue civilians.

Most importantly, it prevents firefighters from being exposed to hazardous conditions while in training.

“I don’t ever see us replacing real fire training,” said Yu, “but this is a means to augment the training to improve the health and safety of the firefighters, as well as make training easier to deploy in a variety of buildings and circumstances. AI allows us to rapidly analyze human performance and then adaptively synthesize training programs to address weakness we observe.”

Martin and Yu are in the early phases of development but hope to deploy their initial training designs with Fairfax County Fire and Rescue soon. They both have hopes for further incorporation of AI and augmented reality technology in firefighting tools, such as headsets that can scan the environment and offer real-time feedback about potential dangers, exit strategies, and health metrics as firefighters enter hazardous zones.

“More and more virtual and immersive reality training is being used [in kinesiology],” said Martin. “I think this project demonstrates the importance of collaboration and interdisciplinary teams to solve challenges in the field.”

Discover how George �鶹��Ƶ is driving AI

In This Story

Craig Yu

Topics

Research

College of Engineering and Computing

College of Education and Human Development

Artificial Intelligence

Community Impact

National Science Foundation

George �鶹��Ƶ’s ROARS lab retains its ranking as a leader in AI safety at the 2024 International Neural Network Verification Competition

Jennifer Pocock — Mon, 19 May 2025 20:47:30 +0000

George �鶹��Ƶ’s ROARS lab retains its ranking as a leader in AI safety at the 2024 International Neural Network Verification Competition Jennifer Pocock Mon, 05/19/2025 - 16:47

In This Story

ThanhVu Nguyen

The team’s NeuralSAT advanced the frontiers of trustworthy artificial intelligence, earning second place in a highly competitive field of international teams advancing the formal verification of AI systems.

Associate Professor ThanhVu Nguyen

NeuralSAT is an open-source verification tool developed by experts in Formal Methods/AI Hai Duong and ThanhVu Nguyen, both of �鶹��Ƶ. It analyzes whether deep neural networks behave as intended, even under adversarial or uncertain conditions. The International Neural Network Verification Competition (VNN-COMP) gathers international research teams working on formal methods for AI safety—methods that are increasingly important as neural networks are deployed in autonomous systems, healthcare, finance, and other mission-critical systems.

"Our goal with NeuralSAT is to bring precision and robustness to AI systems,” said Nguyen, an associate professor in computer science, “especially as they are increasingly used in safety-critical settings."

Hai Duong is a third-year PhD candidate

Each year, the participating VNN-COMP tools are evaluated across a wide set of benchmarks that test scalability, correctness, and performance. The 2024 competition’s results were recently released.

"This year’s benchmarks pushed the boundaries of what verification tools could handle—both in terms of scale and complexity," said Duong, a third-year computer science PhD candidate. "We’re proud that NeuralSAT remained among the top performers, showing strong results across both safety and robustness categories."

NeuralSAT was developed in close collaboration with researchers across multiple institutions, including the University of Virginia’s Matt Dwyer. It builds on recent advances in symbolic reasoning, SAT solving, and abstraction. In its inaugural appearance at VNN-COMP in 2023, NeuralSAT placed fourth overall and received the New Participant Award. In addition to competitions, the tool has been featured at top-tier conferences such as the Symposium on the Foundations of Software Engineering and Computer Aided Verification.

Nguyen, who serves as an organizer and program chair of VNN-COMP 2025, emphasized the significance of the competition in shaping the research landscape.

"The field of neural network verification is rapidly maturing, and VNN-COMP has become the standard for evaluating progress," he said. "Many research papers now compare their tools against VNN-COMP results or use its benchmarks for evaluation. It's exciting to be involved in shaping the competition while also contributing a tool that pushes the research frontier."

David Rosenblum, chair of the Department of Computer Science, was thrilled with ROARS Lab’s achievement. "ThanhVu Nguyen is an established research leader on problems at the intersection of formal methods and deep learning. This team’s outstanding achievement has demonstrated the great potential impact of verification technologies for AI safety," he said.

The long-term goal of NeuralSAT, supported in part by Nguyen’s NSF CAREER Award and Amazon Research Award, is to remain at the forefront of AI verification research while evolving into a practical, scalable, and potentially commercial-grade solution for industry use.

George �鶹��Ƶ is advancing a bold vision for the future of safe and robust AI and is further expanding its leadership with the launch of Virginia's first master’s program in artificial intelligence in fall 2025. This program reflects the university’s deep commitment to cutting-edge education and the development of trustworthy, high-impact AI technologies.

Topics

Artificial Intelligence

National Science Foundation

Computer science; computing; Amazon

computer science

Research

George �鶹��Ƶ launches Virginia’s first public master’s degree in AI

Nathan Kahl — Mon, 31 Mar 2025 14:56:56 +0000

George �鶹��Ƶ launches Virginia’s first public master’s degree in AI Nathan Kahl Mon, 03/31/2025 - 10:56

�鶹��Ƶ will offer a master of science in artificial intelligence (AI) starting this fall, becoming Virginia’s first public university to offer a stand-alone master’s degree in this field.

Photo by Getty Images

Recently approved by the State Council of Higher Education for Virginia (SCHEV), the degree will equip the next generation of AI innovators with a rigorous, interdisciplinary curriculum blending foundational theory with real-world applications, ensuring graduates are prepared to address complex challenges in industry and government.

Amarda Shehu, inaugural vice president and chief AI officer, associate dean for research in the College of Engineering and Computing, and professor in the Department of Computer Science, said, “Seeing this vision come to life has been incredibly rewarding, and I am deeply honored to have led this effort to create a program that will shape the future of AI talent in Virginia and beyond.”

Courses will span core domains such as machine learning foundations and practice; planning and decision-making for intelligent agents; and deep learning fundamentals; providing students with the expertise to build, deploy, and evaluate AI systems across various computing platforms.

Shehu writes and speaks widely on issues around AI. Photo by Office of University Branding

“We had to completely rethink how we teach AI to students,” added Shehu. “Rather than hiding such courses behind long chains of prerequisites, the challenge that we answered is how to design these courses to be largely self-containing and yet offer a rigorous foundation in AI."

Vadim Sokolov, associate professor, Department of Systems Engineering and Operations Research, was one of the leading faculty in the working group designing curriculum. With his mathematics background and his extensive experience teaching the first deep learning course on campus in 2017, Sokolov has great interest in the field and what this means for students. "The idea is students are not just ‘prompt engineers’ but are supposed to understand the basics of the models and be able to tune them and train them for specific tasks.”

Sokolov added that the program is unique in how it enhances this foundation with real-world settings. According to the program, students will be adept in managing the full machine learning operations lifecycle, integrating open-source AI frameworks, and developing secure, scalable AI solutions while effectively collaborating with cross-functional teams and communicating complex AI concepts to diverse stakeholders.

“This program is carefully designed to meet the needs of our community, whether that is government or business, but it also provides a holistic experience to students, from AI ethics to AI policy, and from scalable and secure AI to advanced AI solutions, Sokolov said.

Shehu said, “As AI transforms the way we work, govern, and live, this master’s degree program is more than just a response to demand—it is a commitment to preparing students with both the technical expertise and ethical grounding to shape the future of AI responsibly.”

Discover more AI research at George �鶹��Ƶ

Topics

computer science

Campus News

Artificial Intelligence

Strategic Direction

Nonprofits are in trouble. Could more sensitive chatbots be the answer?

Jennifer Anzaldi — Tue, 18 Mar 2025 14:48:25 +0000

Nonprofits are in trouble. Could more sensitive chatbots be the answer? Jennifer Anzaldi Tue, 03/18/2025 - 10:48

In today’s attention economy, impact-driven organizations are arguably at a disadvantage. Since they have no tangible product to sell, the core of their appeal is emotional rather than practical—the “warm glow” of contributing to a cause you care about. But emotional appeals call for more delicacy and precision than standardized marketing tools, such as mass email campaigns, can sustain. Emotional states vary from person to person—even from moment to moment within the same person.

Photo by Getty Images

Siddharth Bhattacharya and Pallab Sanyal, professors of information systems and operations management at the Donald G. Costello College of Business at �鶹��Ƶ, believe that artificial intelligence (AI) can help solve this problem. A well-designed chatbot could be programmed to calibrate persuasive appeals in real time, delivering messaging more likely to motivate someone to take a desired next step, whether that’s donating money, volunteering time or simply pledging support. Automated solutions, such as chatbots, can be especially rewarding for nonprofits, which tend to be cash-conscious and resource-constrained.

“We completed a project in Minneapolis and are working with other organizations, in Boston, New Jersey and elsewhere, but the focus is always the same,” Sanyal says. “How can we leverage AI to enhance efficiency, reduce costs, and improve service quality in nonprofit organizations?”

Siddarth Bhattacharya. Photo provided

Sanyal and Bhattacharya’s working paper (coauthored by Scott Schanke of University of Wisconsin Milwaukee) describes their recent randomized field experiment with a Minneapolis-based women’s health organization. The researchers designed a custom chatbot to interact with prospective patrons through the organization’s Facebook Messenger app. The bot was programmed to adjust, at random, its responses to be more or less emotional, as well as more or less anthropomorphic (human-like).

“For the anthropomorphic condition, we introduced visual cues such as typing bubbles and slightly delayed response to mimic the experience of messaging with another human,” Sanyal says.

The chatbot’s “emotional” mode featured more subjective, generalizing statements with liberal use of provocative words such as “unfair,” “discrimination” and “unjust.” The “informational” modes leaned more heavily on facts and statistics.

Over the course of hundreds of real Facebook interactions, the moderately emotional chatbot achieved deepest user engagement, as defined by a completed conversation. (Completion rate was critical because after the last interaction, users were redirected to a contact/donation form.) But when the emotional level went from moderate to extreme, more users bailed out on the interaction.

The takeaway may be that “there is a sweet spot where some emotion is important, but beyond that emotions can be bad,” as Bhattacharya explains.

Pallab Sanyal. Photo provided

When human-like features were layered on top of emotionalism, that sweet spot got even smaller. Anthropomorphism lowered completion rates and reduced the organization’s ability to use emotional engagement as a motivational tool.

“In the retail space, studies have shown anthropomorphism to be useful,” Bhattacharya says. “But in a nonprofit context, it’s totally empathy-driven and less transactional. If that is the case, maybe these human cues coming from a bot make people feel creepy, and they back off.”

Sanyal and Bhattacharya say that more customized-chatbot experiments with other nonprofits are in the works. They are taking into careful consideration the success metrics and unique needs of each partner organization.

“Most of the time, we researchers sit in our offices and work on these problems,” Sanyal says. “But one aspect of these projects that I really like is that we are learning so much from talking to these people.”

In collaboration with the organizations concerned, they are designing chatbots that can cater their persuasive appeals more closely to each context and individual interlocutor. If successful, this method would prove that chatbots could become more than a second-best substitute for a salaried human being. They could serve as interactive workshops for crafting and refining an organization’s messaging to a much more granular level than previously possible.

And this would improve the effectiveness of organizational outreach across the board—a consummate example of AI enhancing, rather than displacing, human labor. “This AI is augmenting human functions,” says Sanyal. “It’s not replacing. Sometimes it’s complementing, sometimes it’s supplementing. But at the end of the day, it is just augmenting.”

In This Story

Siddharth Bhattacharya

Pallab Sanyal

Topics

A.I. & Innovation - Costello

ESG - Costello

Costello Research Digital Platforms

Costello Research Artificial Intelligence

Costello Research Machine Learning

Costello College of Business News

Costello College of Business Faculty Research

ISOM Faculty Research

Research

Artificial Intelligence

George �鶹��Ƶ sets in motion a plan to harness AI for responsible adoption and societal impact

Colleen Rich — Mon, 03 Mar 2025 20:07:19 +0000

George �鶹��Ƶ sets in motion a plan to harness AI for responsible adoption and societal impact Colleen Rich Mon, 03/03/2025 - 15:07

At last week’s Board of Visitors meeting, �鶹��Ƶ’s Vice President and Chief AI Officer Amarda Shehu rolled out a new model for universities to advance a responsible approach to harnessing artificial intelligence (AI) and drive societal impact. George �鶹��Ƶ’s model, called AI²Nexus, is building a nexus of collaboration and resources on campus, throughout the region with our vast partnerships, and across the state.

AI²Nexus is based on four key principles: “Integrating AI” to transform education, research, and operations; “Inspiring with AI” to advance higher education and learning for the future workforce; “Innovating with AI” to lead in responsible AI-enabled discovery and advancements across disciplines; and “Impacting with AI” to drive partnerships and community engagement for societal adoption and change.

Shehu said George �鶹��Ƶ can harness its own ecosystem of AI teaching, cutting-edge research, partnerships, and incubators for entrepreneurs to establish a virtuous cycle between foundational and user-inspired AI research within ethical frameworks.

As part of this effort, the university’s AI Task Force, established by President Gregory Washington last year, has developed new guidelines to help the university navigate the rapidly evolving landscape of AI technologies, which are available at gmu.edu/ai-guidelines.

Further, Information Technology Services (ITS) will roll out the NebulaONE academic platform equipping every student, staff, and faculty member with access to hundreds of cutting-edge Generative AI models to support access, performance, and data protection at scale.

“We are anticipating that AI integration will allow us to begin to evaluate and automate some routine processes reducing administrative burdens and freeing up resources for mission-critical activities,” added Charmaine Madison, George �鶹��Ƶ’s vice president of information services and CIO.

George �鶹��Ƶ is already equipping students with AI skills as a leader in developing AI-ready talent ready to compete and new ideas for critical sectors like cybersecurity, public health, and government. In the classroom, the university is developing courses and curriculums to better prepare our students for a rapidly changing world.

In spring 2025, the university launched a cross-disciplinary graduate course, AI: Ethics, Policy, and Society, and in fall 2025, the university is debuting a new undergraduate course open to all students, AI4All: Understanding and Building Artificial Intelligence. A master’s in computer science and machine learning, an Ethics and AI minor for undergraduates of all majors, and a Responsible AI Graduate Certificate are more examples of �鶹��Ƶ’s mission to innovate AI education. New academies are also in development, and the goal is to build an infrastructure of more than 100 active core AI and AI-related courses across George �鶹��Ƶ’s colleges and programs.  

The university will continue to host workshops, conferences, and public forums to shape the discourse on AI ethics and governance while forging deep and meaningful partnerships with industry, government, and community organizations to offer academies to teach and codevelop technologies to meet our global society needs. State Council of Higher Education for Virginia (SCHEV) will partner with the university to host an invite-only George �鶹��Ƶ-SCHEV AI in Education Summit on May 20-21 on the Fairfax Campus.

Virginia Governor Glenn Youngkin has appointed Jamil N. Jaffer, the founder and executive director of the National Security Institute (NSI) at George �鶹��Ƶ’s Antonin Scalia Law School, to the Commonwealth’s new AI Task Force, which will work with legislators to regulate rapidly advancing AI technology.

The university’s AI-in-Government Council is trusted resource for academia, public-sector tech providers, and government for advancing AI approaches, governance frameworks, and robust guardrails to guide AI development and deployment in government.

Learn more about George �鶹��Ƶ’s AI work underway at gmu.edu/AI.

More information on the Board of Visitors' February meeting is forthcoming and will be published in Thursday, March 5, George.

Listen to the episode

Interested in more from Amarda Shehu? Listen to her discuss AI's risks and rewards on the Access to Excellence podcast.

Discover more Artificial Intelligence research from George �鶹��Ƶ

Topics

Campus News

Research

Artificial Intelligence

Strategic Direction

Engineering professors develop human-AI colorization tool for cartoonists

Shayla Brown — Mon, 03 Mar 2025 17:28:34 +0000

Engineering professors develop human-AI colorization tool for cartoonists Shayla Brown Mon, 03/03/2025 - 12:28

�鶹��Ƶ professor Sungsoo Ray Hong has always been a huge fan of cartoons, comics, and animations. By combining his passion with his research area of human-computer interaction, he has created a new tool for cartoonists, ShadowMagic.

Hong, his colleague Yotam Gingold, and their PhD students published a paper, “ShadowMagic: Designing Human-AI Collaborative Support for Comic Professionals’ Shadowing,” elucidating the need for the tool, which aims to boost product quality and designer productivity through study-informed human-AI interaction.

The team worked with professional cartoonists to find common workflow of colorization that is shared within the industry as a reference point for their research, which examines the evolution of designing technology as cartoonists transitioned from using PCs in the 1980s to artificial intelligence, said Hong, who is an assistant professor at the Department of Information Sciences and Technology.

Creating a comic book is a team effort as there are many roles throughout the process including story drafting, sketching, colorizing, editing, and sending to a production company, said Hong.

“Professional cartoonists are sometimes producing content every other day,” said Hong, who is also director of George �鶹��Ƶ’s Alignment Lab. “You cannot commit to these kinds of deadlines doing everything by yourself. That's why the colorization industry is so helpful to professional cartoonists.”

Colorization is a sequential process of line drawing, flatting, base colorization, shadows and/or lighting, background and other special effects. The tool stems from a previous project, FlatMagic, another AI-driven tool designed to improve the creative process for digital comic professionals.

“This is a fairly standardized process, especially in South Korea and growing throughout the United States as well,” he explained. “The solutions to these network visualizations should be contextualized, but we have the limited resources. We can’t build everything for everybody. That’s my motivation here.”

Gingold also specializes in human-computer interaction and “has always been interested in sketching and how computers can help us express ourselves,” said the Department of Computer Science associate professor.

Yotam Gingold (left) and Ray Hong in Research Hall, presenting their colorization tool. Photo by Cristian Torres/Office of University Branding.

“We looked at artistic challenges, usually tedious things that aren’t so fun for the artists and break down those problems, so that they can focus more on the creative parts of it,” said Gingold, who is the director of George �鶹��Ƶ’s Computational Reality, Creativity, and Graphics Lab.

There are millions of pixels in an image, and if an artist wants to move or change the color of a building for example, they point at one pixel among the millions. The artist wants the rest of the pixels to follow along naturally so the rest of the building should also change, he explained.

“The computers will do the math where we can write formulas to update all the other millions of pixels. Another example would be keeping lines parallel and corners perpendicular, and that’s where a lot of my research comes in,” said.

While a doctoral student at George �鶹��Ƶ, Chuan Yan, PhD ’24, was recruited by Gingold for this project and conducted research on sketching and painting. Yan’s area of expertise encompasses computer graphics fundamentals and sketch-related computer vision techniques, which he had learned extensively from Gingold, he explained.

The ShadowMagic tool was built as an Adobe Photoshop plug-in, and since its release, the team “has garnered significant interest from artists inquiring about updates and expressing a strong desire for a tool that could free them from the labor-intensive flatting process,” said Yan, who is currently a postdoc researcher at Stanford University.

“This response is both exciting and humbling. It confirms that our research is addressing a real need in the industry yet also highlights the remaining challenges in making [the tool] ready for widespread adoption.”

Yan believes that as more sophisticated generative AI systems continue to emerge, AI-human collaboration will become increasingly vital.

“I am excited to witness and contribute to this transformation, helping to shape a future where AI acts as a powerful tool that amplifies human creativity rather than replacing it,” he said.

Discover more Artificial Intelligence research from George �鶹��Ƶ

In This Story

Sungsoo Ray Hong

Yotam Gingold

George �鶹��Ƶ alumnus launches AI software startup for law enforcement training

Greg Johnson — Wed, 26 Feb 2025 16:05:48 +0000

George �鶹��Ƶ alumnus launches AI software startup for law enforcement training Greg Johnson Wed, 02/26/2025 - 11:05

Delivering immersive, voice-driven simulations for law enforcement through building skills, gaining real-time feedback, and preparing for real-world scenarios, Kaiden AI is a software startup co-founded by its CEO Ali Aldubaisi, BS Neuroscience ’21.

Ali Aldubaisi

Aldubaisi had begun studying neuroscience at George �鶹��Ƶ with the intention of going to medical school. However, the distance learning forced upon by COVID led him to think outside of the box and start his first company vRealm, an online tutoring platform.

After a few years working on the project, he co-founded Kaiden AI with his vRealm partner, Ali "AJ" Aldjaei. “When we first started, we were focused on the higher education market,” he says. “We initially built the AI platform for grading and evaluation. Like most startups, you build where you initially think is the right fit, and that’s going to be the market for you.” Throughout the process of building his startup and attracting investors, he would find each day bringing surprises.

After seeing limited traction in higher education, Aldubaisi and his team built a small-scale simulation demo to test a new direction. Among the first to see it was Steve Gladis, a senior scholar at George �鶹��Ƶ’s Center for the Advancement of Well-Being and a former FBI special agent. He immediately recognized its potential for law enforcement training and encouraged Aldubaisi to explore that path.

“We put the MVP [minimum viable product] in front of police chiefs and academy directors, and their reaction was completely different from our earlier demos,” says Aldubaisi. “That’s when we knew we had to double down.”

Now, Aldubaisi and his team partner with law enforcement agencies and regional academies to implement their simulation platform, with organizations like the Northern Virginia Criminal Justice Academy currently piloting the technology. Recently closing a $1 million pre-seed round, Kaiden AI’s investors are venture capitalists largely outside of the law enforcement space.

Ali Aldubaisi (right) at the 2024 Patriot Pitch competition. Photo by John Boal Photography

Even though his current work does not appear directly related to neuroscience, Aldubaisi recognizes the value of his degree. In fact, he still keeps up with neuroscience journals, seeing the relevance between that field and running his company. Developing hypotheses and constantly experimenting has helped Aldubaisi and his team refine and optimize Kaiden AI, from its initial concept in January 2024 to its current phase today.

“A lot of the concepts mirror each other,” he says. “What goes into marketing, how to understand people, how people learn.”

In 2024, Aldubaisi led the keynote discussion at the Patriot Pitch Competition, an event organized by the Center for Innovation and Entrepreneurship to support ideas and early-stage business ventures from George �鶹��Ƶ students and recent alumni. He is now in talks with Rebecca Howick, the center’s director of operations and engagement, about getting more involved with their efforts.

“Having a center where somebody in whatever discipline that they’re studying can articulate or put together an idea or experiment on ideas is so crucial,” says Aldubaisi. He knows there are many intelligent people with brilliant ideas who may be hesitant to pursue their ventures because they don’t know how they’re going to get the funding or backing.

Having gone through that process, Aldubaisi offers crucial advice to aspiring entrepreneurs that reflects his scientific background. “Having the courage to focus on that one thing that you’re trying to prove out is probably the biggest thing,” he says. “Second thing is always look at what your idea is as an experiment, and the goal is to prove or disprove as soon as possible.”

By following these guidelines, Aldubaisi and his team have surpassed fundraising milestones and are excited to put Kaiden AI into the hands of many more law enforcement students and officers.

In This Story

Rebecca Howick

Learn more about AI research at George �鶹��Ƶ

Topics

Costello College of Business News

Center for Innovation and Entrepreneurship

entrepreneurship

Center for Advancement of Well-Being

Artificial Intelligence

Alumni

Keeping data safe by keeping it separate

Teresa Donnellan — Tue, 25 Feb 2025 15:35:36 +0000

Keeping data safe by keeping it separate Teresa Donnellan Tue, 02/25/2025 - 10:35

Developing a robust cybersecurity system demands a lot of data, but keeping all that data in a single location introduces the risk of a data breach. Luckily, �鶹��Ƶ researcher Zhuangdi Zhu in the Department of Cyber Security Engineering specializes in federated learning, a method of decentralized machine learning that allows multiple entities to collaboratively train a machine learning model without sharing their raw data.

Zhu is using this expertise to assist Wajih Ul Hassan of the University of Virginia develop scalable host-based intrusion detection systems (HIDS). Together, Zhu and Hassan are leveraging artificial intelligence (AI) to enhance a given host’s cybersecurity while ensuring data privacy.

HIDS monitor and analyze the activities within a single host or device. These systems use data, such as system logs, web browsing logs, and process logs, to detect irregularities that may indicate a security threat. For example, Zhu noted, a HIDS might detect malicious software that prompts a browser to download a harmful PDF file. By identifying such irregularities, the HIDS can prevent potential security breaches. Central management of these logs through managed security service providers (MSSPs) on cloud servers, however, could potentially breach user privacy. Enter: federated learning.

Zhuangdi Zhu, photo provided

In a federated learning system, each entity trains a local model on its own data and only shares the model updates with a central server. Local models are trained on individual datasets, and their findings are sent to a central server, Zhu explained. The server then aggregates these parameters to create a global model, which is subsequently sent back to the local models. This iterative process continues until a robust and accurate model is developed. Federated learning thus ensures sensitive data remains within the local models, addressing privacy concerns and complying with data regulations.

The project, funded by the Commonwealth Cyber Initiative, is set to span one year, starting in the summer of 2025. In addition to developing a federated learning framework for HIDS, the team plans to create an open-source dataset for intrusion detection, which will serve as a valuable resource for future research and development in this field.

Zhu's work on decentralized machine learning not only addresses critical privacy concerns but also paves the way for more secure and efficient threat detection systems. Her work on this project represents a significant step forward in the ongoing effort to protect digital environments from evolving cyber threats.

In This Story

Zhuangdi Zhu

Learn more about George �鶹��Ƶ's Department of Cyber Security Engineering

Topics

Cybersecurity

Commonwealth Cyber Initiative (CCI)

Machine Learning

Artificial Intelligence

Research

Artificial Intelligence

George �鶹��Ƶ launches its inaugural ACCESS Academy cohort

Powering Artificial Intelligence at George �鶹��Ƶ

Learn more about ACCESS Academy

Related News

Topics

AI4Defense develops DoD-driven smart solutions

In This Story

Learn more about Research at George �鶹��Ƶ

Related Stories

Topics

Trial by virtual fire: Augmented reality project tailors training for firefighters

Discover how George �鶹��Ƶ is driving AI

In This Story

Related News

Topics

George �鶹��Ƶ’s ROARS lab retains its ranking as a leader in AI safety at the 2024 International Neural Network Verification Competition

In This Story

Topics

George �鶹��Ƶ launches Virginia’s first public master’s degree in AI

Discover more AI research at George �鶹��Ƶ

Related News

Topics

Nonprofits are in trouble. Could more sensitive chatbots be the answer?

In This Story

Related Stories

Topics

George �鶹��Ƶ sets in motion a plan to harness AI for responsible adoption and societal impact

Discover more Artificial Intelligence research from George �鶹��Ƶ

Related News

Topics

Engineering professors develop human-AI colorization tool for cartoonists

Discover more Artificial Intelligence research from George �鶹��Ƶ

In This Story

Related News

More from �鶹��Ƶ Spirit Magazine

George �鶹��Ƶ alumnus launches AI software startup for law enforcement training

In This Story

Learn more about AI research at George �鶹��Ƶ

Related News

Topics

Keeping data safe by keeping it separate

In This Story

Learn more about George �鶹��Ƶ's Department of Cyber Security Engineering

Related Stories

Topics