AI Drugmaker Targets Obesity Beyond GLP-1

Insilico advances oral GIP candidate, signalling wider competition in weight-loss treatments  

16 Feb 2026

The race to develop new obesity medicines is widening beyond established GLP-1 therapies, as Insilico Medicine nominates an artificial intelligence-designed drug as a preclinical candidate.

The company said its oral small-molecule therapy, ISM0686, targets the GIP receptor, a pathway linked to metabolism and energy balance. The drug has shown efficacy in mouse models but has not yet been tested in humans.

Unlike leading GLP-1 treatments, which are largely delivered by injection, an oral alternative could expand patient access and convenience if clinical trials confirm its safety and effectiveness. Several pharmaceutical groups are exploring similar approaches as demand for weight-loss drugs accelerates.

Insilico said it moved from concept to preclinical nomination in about 14 months, using AI tools to design and refine the compound. Fewer than 200 molecules were synthesised during development, a lower number than in many traditional drug discovery programmes.

In mouse studies, ISM0686 produced about 10 per cent weight loss as a standalone therapy and up to 31 per cent when combined with semaglutide, a widely used GLP-1 treatment. The results remain preliminary and are limited to animal models.

The commercial stakes are high. Industry forecasts suggest the global market for obesity treatments could reach about $60bn by 2030, as rising prevalence and expanding reimbursement drive demand. The prospect of sustained growth has intensified competition among biotechnology companies and large pharmaceutical groups seeking new mechanisms, combination regimens and oral formulations.

Regulators are expected to scrutinise long-term safety and durability of effect as more candidates move into clinical testing. Historically, promising animal data have not always translated into success in human trials.

For Insilico, the programme also serves as a test of AI-driven drug development. The company argues that machine learning can shorten timelines and reduce laboratory work in early discovery. Whether that approach can deliver approved therapies at scale will depend on the outcome of future clinical studies.