i-GENTIC launches AI agents to boost FDA 510(k) success

i-GENTIC AI has launched a suite of artificial intelligence agents designed to reduce rework in US Food and Drug Administration 510(k) submissions, a common route for medical devices to reach the US market.

Its context-aware agents review submission packages for internal consistency across documents and data sets, a process it compares to the scrutiny applied by FDA reviewers. The product is built around i-GENTIC's GENIE reasoning engine.

The 510(k) pathway requires companies to show a device is substantially equivalent to a legally marketed predicate. Manufacturers typically assemble large documentation sets covering intended use, risk controls, verification and validation testing, and labelling. When information does not align across these materials, regulators often issue requests for additional information, extending review cycles.

i-GENTIC says the most common problem is not missing documents, but inconsistencies introduced after design changes and internal reviews. It cited FDA-reported figures showing 76% of 510(k) submissions fail in the first round.

The agents check whether claims remain supported by test evidence and whether risk controls still map to expected device behaviour. They also look for drift between intended use statements and the design assumptions that informed risk assessments and performance testing.

"FDA reviewers look for a consistent narrative," said Zahra Timsah, founder and CEO of i-GENTIC AI. "They need to see that evidence, logic, and traceability for your device remain solid across the entire submission."

Internal data from i-GENTIC suggests first-time submission success rates improve by up to 97% when teams use its agents, and approval timelines drop by an average of six weeks. The company did not disclose how many projects underpin those figures or what types of devices were involved.

How it works

Unlike tools that focus on document completeness, i-GENTIC positions the agents as a way to detect contradictions across an entire submission package. This includes cases where each document appears correct in isolation, but the overall logic breaks down when read together.

The system includes context awareness and multilingual support, which i-GENTIC says allows it to work across diverse documentation sources and writing styles. The company has not specified which languages are supported or whether the agents integrate with particular quality management systems, document repositories, or product lifecycle platforms.

"Traditional systems act like a basic checklist, telling you that a document exists. But it can't tell you if your data still makes sense after months of design changes and internal reviews," said Ken Washington.

Washington, a former chief technology officer at Medtronic, is chairman of i-GENTIC's board. The company highlighted his experience overseeing regulatory and technology strategy across medical device portfolios.

"As device portfolios scale, complexity grows faster than teams can manually track across systems. Context-aware agents help surface weak links early, when teams still have time to fix them thoughtfully rather than discovering them months later through regulatory back-and-forth," he said.

Case example

i-GENTIC described a recent engagement with a medical device company preparing a 510(k) submission. It said the company had completed the required documentation and passed internal checklist reviews.

The agents flagged a mismatch after the intended use statement was updated late in development, while the risk analysis and performance testing rationale still reflected earlier assumptions. i-GENTIC traced the inconsistency across the Design History File, risk management file, and test reports. The customer corrected the issue before filing, reducing the likelihood of questions during review.

Scott Van Valkenburgh, an i-GENTIC board member and senior vice president at Appian, said the tool reflects a broader shift in compliance automation for medtech firms.

"Speed to clearance is a major competitive advantage. Traditional AI tools are no longer enough for the rigors of FDA scrutiny," Van Valkenburgh said. "Regulations are in place to make sure devices are safe, and context-aware agents can help MedTechs cross the 510(k) finish line ahead of their competitors."

Beyond medtech

i-GENTIC is initially focusing on medical devices and the 510(k) process, but says the same approach can apply in other regulated industries that require consistency and traceability across policies, standard operating procedures, and governance frameworks.

i-GENTIC is part of AdvaMed, the medical technology association. The company positions its GENIE platform as an orchestration layer that coordinates specialised agents across workflows, documents, and systems.

It also plans to expand the platform's use cases beyond 510(k) submissions as more compliance teams seek automated ways to reconcile changes across complex documentation sets.