TechM targets GenAI-enabled MVPs for life sciences

Does anyone else see new lipstick on an old pig? That’s how the buzz around generative artificial intelligence (GenAI) seems when it is being applied to solve legacy challenges without addressing the causes. Yet, some, like TechM, are quietly and diligently developing GenAI-enabled minimum viable products (MVP). TechM’s Generative AI Studio is part of its amplifAI offerings suite for life-sciences-specific challenges that could help clients explore new ways to use GenAI to solve their problems and create solutions.

Life sciences like pharma, medtech, and contract research organizations are under increased pressure to optimize their costs as price negotiation ramps up, accelerate drug development, and go to market to meet consumer expectations—all this, and yet remain compliant with regulations. So, there is an intrinsic desire to explore using GenAI to address the headwinds. The life sciences have generally been ahead of the market when it comes to adopting new technologies (see Exhibit 1), giving TechM the impetus to drive forward with its amplifAI suite.

Exhibit 1: Life sciences will likely apply GenAI at scale faster than other industries, based on past behaviors

Sample: HFS OneOffice Pulse Study, H1 2021; 800 respondents from Global 2000 enterprises
Source: HFS Research, 2023

Tech Mahindra offers GenAI experiment-as-a-service to clients for exploring MVPs—a near-term practical win

TechM is partnering with life science enterprises to develop proofs of concept to promote minimum viable products. The MVPs are TechM’s candidates to help clients scale new GenAI project investments. While it is too early to advertise real outcomes, TechM is being deliberate about what “good” looks like at this stage of GenAI experimentation. An early example of a GenAI experiment-as-a-service implementation explores using GenAI to help manage the pharmacovigilance (PV) process.

Life sciences enterprises have incorporated several processes to comply with various regulatory requirements. A prime example is the many activities to address PV. While there has been technology enablement to digitize and automate various PV activities, many remain manual and cumbersome. Enabling GenAI to summarize adverse event reports to improve signal detection and safety writing services can significantly improve efficiency and regulatory compliance.

There are many possibilities to apply GenAI throughout life sciences, but not all make sense. In the PV example and others, TechM uses its Generative AI Studio, an experimentation studio agnostic of the many large language models (LLM). It has incorporated six content modalities, including images, documents, video, and data. The studio supports both open-source and proprietary models. It is also cloud agnostic, making the studio very flexible given the level of experimentation occurring. It has 50+ in-built capabilities for clients to experiment with across 10+ model families.

Apart from experimentation, the studio acts as a test workbench, and it helps create a prompt repository for enterprises. It also serves as an ideation tool, providing a prescriptive architecture for creating enterprise AI solutions.

The GenAI toolset can provide critical go/no-go decisions for TechM and its clients, helping assess whether GenAI is a viable and appropriate tool.

The Bottom Line: It is critical to address the right type of problem with GenAI, or else it’s like racing a Lamborghini in rush hour traffic.

Life sciences leaders may have a generational opportunity to reimagine their drug discovery protocols, reconfigure their supply chains, and potentially get closer to their consumers by enabling GenAI. While GenAI is not the answer to all questions, it certainly opens the ability to enable business processes with increased intelligence and automation that may not have been possible in the past. So, when the buzz of GenAI subsides, it is definitely still prudent to keep exploring the possibilities.