Isomorphic Labs released a technical report touting IsoDDE, a drug‑discovery model its scientists say advances prediction of protein–drug interactions and antibody structures, but the company is keeping the model proprietary. The release drew comparisons to AlphaFold’s earlier leaps; researchers cited in Nature called IsoDDE a potential step on par with an "AlphaFold 4." Academics and open‑source developers flagged the lack of methodological transparency as a barrier to replication and community progress. The technical paper provides performance claims but few implementation details, leaving groups developing open models—such as Boltz‑2—uncertain how to match IsoDDE’s results. The move highlights a growing tension between private AI drug‑discovery commercialization and the open science community’s drive for reproducibility and shared tooling.