TCS launches AgentHub for drug development AI tasks

TCS has launched TCS ADD AgentHub for drug development, aimed at pharmaceutical companies working in regulated research and safety functions.

The platform is designed to let drugmakers deploy artificial intelligence agents across clinical development and pharmacovigilance while maintaining human oversight and audit controls. Built on the TCS ADD framework, it can be tailored to fit customer workflows.

Pharmaceutical groups have been testing AI across research and development, but adoption has been uneven because of governance concerns, fragmented data systems and stricter regulatory expectations. TCS is targeting those constraints with a model that assigns defined roles to AI agents and embeds oversight into day-to-day processes.

The platform supports a range of tasks in clinical development and drug safety, including ICSR intake, data entry, coding, review, literature analysis, study design, protocol digitisation, clinical data review, SDTM transformation and medical monitoring assistance.

Customers can roll out agents progressively rather than replace existing systems in a single step. The platform is also designed for rapid integration with existing clinical and safety operations.

Operational focus

The launch reflects wider demand from pharmaceutical companies for tools that can handle larger volumes of research and safety data without weakening compliance standards. Clinical trials and adverse event monitoring generate large, often complex datasets, and companies are under pressure to process them faster while maintaining clear records for regulators.

TCS said the platform has shown measurable effects across several parts of the research chain. According to the company, applications built on the system delivered up to 40% efficiency gains in clinical data management, up to 30% lower clinical study build effort through metadata-driven automation, and up to 30% cost savings in end-to-end safety case processing.

TCS also said AI safety agents could reduce quality control effort by as much as 50%. If replicated across broader deployments, those figures would place the product in a market where service providers and drugmakers are both seeking to cut repetitive manual work in heavily documented processes.

Human oversight

TCS described the platform's operating model as one in which AI agents work within enterprise workflows while people retain responsibility for governance and decision-making. That approach is becoming more common in regulated sectors, where companies want automation but remain cautious about handing critical judgement entirely to software.

Debashis Ghosh, president, lifesciences and healthcare, TCS, outlined the company's position on the launch.

"TCS ADD AgentHub is a role-based, enterprise-ready and trusted AI platform that will enable our customers to accelerate drug development using agentic AI at scale. It enables a shift from reactive to proactive, scalable and audit-ready operations amid an ever-changing regulatory environment. TCS' strategy is to move towards autonomous enterprise functions where an agentic AI workforce operates alongside humans, driving innovation in drug development and improving patient safety," said Debashis Ghosh, president, lifesciences and healthcare, TCS.

Sector push

The move adds to a growing race among technology suppliers to build AI systems for life sciences that fit within existing compliance structures rather than sit outside them. Drug development has become a key target because it combines high operating costs, long timelines and extensive documentation requirements, making it attractive for automation if companies can satisfy regulators and internal risk teams.

TCS is positioning the product across the research and development value chain rather than within a single narrow use case. That broad approach may appeal to larger pharmaceutical groups seeking one framework for multiple functions, from trial design and clinical data handling to pharmacovigilance and safety review.

TCS said its catalogue of AI agents will evolve to match customer needs and operating environments. The agents can be deployed progressively with limited implementation effort, allowing organisations to standardise AI use across processes while shifting scientific teams towards higher-value work.