SciBridge AI toolkit aids FDA GRAS submissions

SciPinion has launched SciBridge, an AI-enabled software suite designed to help food producers quickly demonstrate that their ingredients are substantially similar to FDA-cleared GRAS substances. The SciBridge Safety Suite compares a candidate food ingredient against substances reviewed through the FDA’s GRAS notice program, turning months of manual work into a documented report in about ten minutes.

The suite is built from the FDA’s GRAS notice inventory and combines AI analysis with human expert review. Its three core components are Browse, Screen, and Bridge. The underlying tools were shaped over nearly a year of expert-led quality control and prompt refinement.

Tool addresses looming mandatory GRAS notification under proposed FDA rule

SciBridge is positioned to help companies handle a proposed FDA rule that would end self-affirmed GRAS status and require formal notification. The FDA published this proposal on August 11. Public comments are due by December 9.

If passed, the new rule would mandate that food producers submit GRAS notices to the FDA. Only narrow exemptions would apply, specifically for ingredients shown to be the same as one the FDA has already reviewed with a 'no questions' letter. Exemptions only apply if manufacturing and intended applications are substantially equivalent to previously reviewed conditions. The proposed rule does not define a standard method for showing this similarity.

Exemption pathway could affect tens of thousands of ingredients, per SciPinion

SciPinion states the scale of companies exposed to this exemption is far larger than previously thought. The firm estimates potentially tens of thousands of 'substitutional' ingredients rely on equivalence to a single precedent notice. These are cases where multiple manufacturers rely on the same single notice. SciPinion acknowledged that no public data exists to confirm this estimate.

Precision-fermented proteins, mycoprotein, and plant protein isolates have often reached US shelves on self-affirmed GRAS conclusions that were never filed with the agency. The new proposal would change that.

SciBridge built with expert input and data security for industry use

The AI toolkit is built for startups, ingredient and finished goods manufacturers and distributors, trade associations, regulatory consultants, and law firms. Development was informed by FDA feedback, though SciPinion clarifies this does not amount to FDA review or endorsement. Vanessa Feher Castagna, PhD, is Director of Food Safety & Claims at SciPinion. Dennis Keefe, PhD, former Director of FDA’s Office of Food Additive Safety, also contributed.

Data security was built as a core requirement to protect confidential ingredient information. The software is available by subscription or through professional services. SciPinion also offers GRAS consulting and portfolio assessments for companies moving from self-affirmed to notified status.

Early adopters include TurtleTree as SciPinion targets food-tech market

Early users include TurtleTree, a Novonesis-backed precision-fermented lactoferrin startup that received a 'no questions' letter last year. Aletta Schnitzler is chief scientific officer at TurtleTree. SciBridge aims to give food-tech firms a faster path to market under the evolving GRAS rules, which currently allow ingredients to launch without full pre-market approval.

SciBridge is available by subscription or through professional services as companies prepare for the December 9 comment deadline on the proposed GRAS rule.