Boomi, LP recently secured a major technological breakthrough with the issuance of U.S. Patent No. US12777064B2. This patent, titled ‘Artificial-intelligence-based proactive cybersecurity for integration processes’, introduces a powerful way to protect complex workflow automations from digital threats. Developers and engineers can now identify vulnerabilities early in the workflow creation cycle rather than relying on reactive security scans after deployment.
PENNSYLVANIA PATENT OF THE MONTH – NOVEMBER 2026
The innovation focuses on proactive threat mitigation within integration platforms, addressing security risks common in low-code and enterprise integration development. Invented by Manpreet Singh, Tyeshia Pegram Fornville, and Michael Bachman, this architecture employs generative artificial intelligence combined with crowdsourced integration configurations. By generating synthetic data targeted at specific security endpoints during integration testing, the system proactively detects and resolves vulnerabilities before any integration process is pushed into production.
Award Recognition by Swanson Reed
Swanson Reed selected this technology as the Pennsylvania Patent of the Month for November 2026 due to its inventive solution to enterprise cybersecurity challenges. As businesses increasingly rely on distributed software integrations and low-code builders, the risk of deploying unvetted configurations escalates dramatically. Swanson Reed recognized the patent for its clear technical novelty, addressing industry uncertainty surrounding automated pre-deployment testing and significantly advancing the state of artificial intelligence in platform security.
R&D Tax Credit Eligibility
Developing and refining the practical applications of this patent presents clear eligibility for the federal Research and Development (R&D) Tax Credit under Internal Revenue Code Section 41. Companies that design, train, and test custom generative AI models to simulate cyber attacks on integration pipelines must systematically resolve technical uncertainties. The engineering process entails formulating hypothesis-driven testing environments, building algorithm pipelines to generate domain-specific synthetic threat data, and iteratively tuning models to minimize false detection rates. Qualified research expenses, such as the wages of data scientists, machine learning engineers, and software architects, along with cloud computing costs used during model development and testing, can be claimed to substantially reduce federal tax liability.
