Secure AI Hackathon Concludes with Innovations to Strengthen AI Security
Tag: General news
Published On: October 08, 2026
The Secure AI Hackathon, an initiative under the Cybersecurity and Artificial Intelligence Research Lab (CAIRLab KNUST) of the College of Science, KNUST has concluded its maiden edition with participants developing practical solutions to emerging privacy and security risks associated with artificial intelligence.
Led by Dr. Emmanuel Ahene (Ph.D), the CAIR Lab organised the hackathon over several weeks, bringing together students, developers, researchers and technology enthusiasts from Ghana and beyond to develop secure, reliable and user-friendly AI solutions.
The initiative sought to help users identify and mitigate privacy and security risks when interacting with AI tools, while encouraging participants to translate their technical knowledge into solutions with practical applications.
The hackathon culminated in final-day presentations by 13 teams, whose solutions were assessed based on factors including practicality, impact and reliability.
Dr. Ahene said the growing use of AI in areas such as healthcare, banking, education and public services made it necessary to consider security and privacy alongside functionality.
“AI is moving from the labs to hospitals, banks, classrooms and even government services. We therefore need to look at how we can make it safer, especially in our working environments and critical infrastructure,” he said.
He emphasized that secure AI development should go beyond creating systems that function to ensuring that models are protected against different forms of attacks, safeguard users’ privacy and are safe for deployment.
The hackathon was designed not merely as an academic competition but as a platform for developing products that could potentially serve industry, government institutions and other users.
“We are not here just to give out money. We want you to give us a product that becomes very useful for the security domain; a product that we can recommend to organisations, government services or institutions,” Dr. Ahene said.
Team Zerooth emerged as the overall winner with Guard+, an additional security layer designed to secure AI assistants beyond conventional filters. The team received US$700.
Team Athena Tecne placed second with Athena Shield, a defence-in-depth solution focused on securing AI conversations, receiving US$200.
Team Neuralynx finished third with Ohwefo, a Secure AI Guard designed to understand Ghanaian contexts. The team said “Ohwefo”, a Twi word meaning caretaker, reflects the solution’s role as a protective layer for AI users. It received US$100.
The CAIR Lab plans to continue mentoring participants and supporting the refinement of their solutions into viable products for industry while it also plan to make the Secure AI Hackathon an annual initiative, creating a continuing platform for students, researchers and developers to build practical technologies for a safer and more responsible AI ecosystem.
Led by Dr. Emmanuel Ahene (Ph.D), the CAIR Lab organised the hackathon over several weeks, bringing together students, developers, researchers and technology enthusiasts from Ghana and beyond to develop secure, reliable and user-friendly AI solutions.
The initiative sought to help users identify and mitigate privacy and security risks when interacting with AI tools, while encouraging participants to translate their technical knowledge into solutions with practical applications.
The hackathon culminated in final-day presentations by 13 teams, whose solutions were assessed based on factors including practicality, impact and reliability.
Dr. Ahene said the growing use of AI in areas such as healthcare, banking, education and public services made it necessary to consider security and privacy alongside functionality.
“AI is moving from the labs to hospitals, banks, classrooms and even government services. We therefore need to look at how we can make it safer, especially in our working environments and critical infrastructure,” he said.
He emphasized that secure AI development should go beyond creating systems that function to ensuring that models are protected against different forms of attacks, safeguard users’ privacy and are safe for deployment.
The hackathon was designed not merely as an academic competition but as a platform for developing products that could potentially serve industry, government institutions and other users.
“We are not here just to give out money. We want you to give us a product that becomes very useful for the security domain; a product that we can recommend to organisations, government services or institutions,” Dr. Ahene said.
Team Zerooth emerged as the overall winner with Guard+, an additional security layer designed to secure AI assistants beyond conventional filters. The team received US$700.
Team Athena Tecne placed second with Athena Shield, a defence-in-depth solution focused on securing AI conversations, receiving US$200.
Team Neuralynx finished third with Ohwefo, a Secure AI Guard designed to understand Ghanaian contexts. The team said “Ohwefo”, a Twi word meaning caretaker, reflects the solution’s role as a protective layer for AI users. It received US$100.
The CAIR Lab plans to continue mentoring participants and supporting the refinement of their solutions into viable products for industry while it also plan to make the Secure AI Hackathon an annual initiative, creating a continuing platform for students, researchers and developers to build practical technologies for a safer and more responsible AI ecosystem.