Drug development is a complex and time-consuming enterprise that has traditionally relied on the expertise of scientists and trial-and-error experimentation. The emergence of artificial intelligence (AI) is beginning to change how researchers approach this process. Across the drug development pipeline, from early target discovery to clinical trial optimization to post-market surveillance, AI-driven methodologies are already delivering subtle but meaningful improvements in speed, efficiency, and decision making.

Despite these strides, many AI tools struggle to meet the specific needs of life sciences research and development (R&D). General-purpose models often lack the scientific context, data explainability, and reasoning capabilities required for high-stakes decisions in biomedical research. That’s why a growing number of scientists are turning to domain-specific platforms built from the ground up for hypothesis generation, causal reasoning, and biological insight.

To explore what this shift means in practice, Drug Discovery News spoke with Yiannis Kiachopoulos, co-founder and Chief Executive Officer (CEO) of Causaly. He shared how their platform is helping top pharmaceutical companies uncover novel targets and make more confident early discovery decisions — faster than ever before.

For those unfamiliar, what exactly does Causaly do and how does it help scientists in drug discovery?

Causaly is a life sciences AI platform that enables scientists to find, interpret, and share biomedical information with unprecedented precision and efficiency. Our suite of AI applications uncovers novel targets, biomarkers, and mechanisms so that scientists can generate and validate hypotheses four times faster. Our proprietary AI is designed specifically for life sciences and has been honed for over eight years. We have also built multiple knowledge graphs that, when combined, scan over 500 million data points and surface crucial cause-and-effect relationships in seconds.

You’ve said that life sciences has been underserved by AI tools. What specific gaps or frustrations in R&D workflows inspired the creation of Causaly?

The way R&D works today is too slow, expensive, and fragmented. The average cost of drug development rose in 2024 to $2.2 billion. Validating a target can take months, and drug success rates are persistently low. These decisions are often caught late and cost life sciences organizations billions.

Life science leaders recognize the need for AI, but they are now faced with the difficult build vs. buy dilemma. Building in-house can offer control and tailored solutions, but it takes years to design an AI with that level of nuance and reliability, which is far too long. Our production-grade AI platform integrates easily across global research organizations thanks to a flexible architecture that adapts to proprietary data and diverse workflows. This means that R&D teams don’t have to choose between speed, effectiveness, or scale. They get all of it with Causaly.

Causaly positions itself not as a search tool, but as a foundation for reasoning. Can you explain how your platform enables scientists to go beyond traditional literature search and reach novel insights?

Traditional search tools were built to retrieve information. Causaly was built to reason with it and critically evaluate the science.

In today’s fast-paced and complex R&D landscape, scientists aren’t just looking for facts — they’re looking for answers, connections, insights, and new directions. That’s why we’ve moved beyond conventional search and built a foundation for scientific reasoning. Our platform can interpret data, generate hypotheses, and uncover insights scientists might otherwise overlook.

Causaly’s scientific AI agents can:

• Interpret both public and internal scientific data
• Map complex biological relationships across genes, diseases, pathways, and more
• Find connections between multiple studies that would be difficult or impossible to spot manually
• Generate novel, evidence-linked hypotheses that are fully transparent and explainable
• Point out limitations in the data so scientists can flag evidence that may be weak or contradictory

Our agents act like scientific collaborators to proactively suggest ideas and identify overlooked connections. The result is a significant reduction in time-to-discovery and a powerful shift in how scientists interact with information.

Many AI tools are now claiming to assist with drug discovery. How do you respond to concerns that there’s a lot of noise in the market? What makes Causaly’s outcomes more reliable or valuable?

Not all AI is created equal, and not every tool is designed for the unique demands of life sciences R&D.

Our scientific AI platform is not retrofitted from general-purpose or business-to-consumer (B2C) tools. This singular focus makes a difference. Our platform is grounded in deep scientific rigor, validated by an in-house team of domain experts, and trusted by the world’s leading pharmaceutical companies for critical R&D workflows.

What sets Causaly apart is the reliability and depth of its outputs:

• 98 percent accuracy in drug-disease relationships
• 96 percent accuracy in drug-target relationships
• A proprietary biomedical knowledge graph with more than 500 million data points, growing by four million each month
• A multi-method AI architecture that scans, validates, and synthesizes information using three complementary reasoning engines, while most tools rely on just one or two

Unlike many newer entrants, Causaly isn’t a generalist platform making broad claims. We’re a scientific AI platform built for one purpose: to help R&D scientists accelerate breakthroughs with confidence and precision. Our AI doesn’t just retrieve data, it helps researchers understand complex biological relationships, form testable hypotheses, and make better decisions faster.

Causaly is now used by several top 50 pharma companies. How are they using the platform today, and what impact has it had on target discovery or portfolio decisions?

Top pharmaceutical organizations are using Causaly to accelerate some of the most critical and resource-intensive stages of early drug discovery — particularly in target identification and prioritization, biomarker discovery, and understanding disease pathophysiology and safety profiling.

In these workflows, scientists face an overwhelming volume of unstructured information scattered across publications, databases, and internal reports. Causaly makes that information instantly searchable, explainable, and actionable by mapping complex biological relationships visually so that researchers can quickly generate and evaluate testable hypotheses.

The impact has been significant. One customer reported that what previously took four weeks for a target prioritization exercise can now be completed in just five days using Causaly’s platform. Others consistently cite our ability to surface unexpected findings, speed up evidence review, and improve confidence in early-stage decisions.

In the case of ProQR Therapeutics, for example, Causaly was instrumental in supporting rapid target identification by helping the team quickly navigate complex disease biology and surface strong, evidence-based rationales for new research directions. They met their 2024 target identification goal by Q3.

What would you say to an R&D leader who’s skeptical that AI can meaningfully support scientific reasoning and hypothesis generation?

I’d say their skepticism is completely valid and, in fact, shared by many of the R&D leaders we work with. The truth is that most AI tools in the market today don’t live up to this promise. They lack scientific transparency and often fall short when applied to the complexity of biomedical research. With these gaps, the tools do not deliver value to scientists and, as a result, do not get adopted.

In comparison, Causaly supports scientific reasoning in a way that’s transparent, explainable, and grounded in evidence. Our AI isn’t a replacement for scientific thinking — it acts as a partner, helping researchers ask better questions, identifying hidden relationships in vast datasets, and generating testable hypotheses supported by clear, traceable evidence.

AI can't replace the scientific process, but when designed properly, it can emulate and enhance it. That’s what Causaly does. We’re quickly becoming a trusted partner in discovery, enabling scientists to focus their expertise on where it matters most.

You’re positioning Causaly as foundational infrastructure — not a one-off tool. What does that mean for how you scale and evolve the platform over the next few years?

It means that we’re not just solving one problem or offering a point solution — we’re building the core intelligence layer that powers scientific reasoning across the entire R&D value chain. Our goal is to fundamentally change how enterprise R&D teams generate hypotheses, make decisions, and scale innovation across therapeutic areas and portfolios.

This long-term view shapes everything about how we build and scale the platform:

• Deep integration into scientific workflows: We’re focused on embedding Causaly into the daily processes of research teams, not sitting alongside them. That includes connecting to internal data sources, interoperating with existing tools, and enabling collaboration across functions like biology, informatics, and translational science.
• Scalable AI architecture: Our agentic AI is built to grow with the needs of enterprise pharma, reason across larger, more complex datasets, and support increasingly sophisticated questions over time. We’ve designed our platform to adapt with our customers, whether they’re centralizing target ID across a global organization or accelerating therapeutic strategy at the portfolio level.
• Domain specificity and scientific depth: We’re investing in continuously expanding the Causaly Knowledge Graph, enhancing scientific reasoning capabilities, and launching offerings like Causaly Deep Research — a new layer of AI designed to emulate the scientific method itself, with testable hypotheses and transparent assumptions.
• Enterprise-grade scale and reliability: Foundational infrastructure needs to be trusted, secure, and scalable across global organizations. That’s why we prioritize performance, data governance, and return of investment measurement, helping customers track impact at the program, team, and portfolio levels.

Given your recent meeting at 10 Downing Street, how do you view the UK’s role in shaping the global AI and life sciences innovation ecosystem?

The UK has a unique opportunity to lead at the intersection of AI and life sciences. It combines world-class research institutions, a thriving biotech sector, deep regulatory expertise, and a growing AI talent base. At Causaly, we see the UK not just as a hub of innovation, but as a proving ground for how AI can be applied responsibly and effectively in scientific domains that truly matter — like drug discovery, public health, and translational research.

Our recent meeting at 10 Downing Street reinforced that there is clear momentum around AI policy and investment, but also a recognition that real impact comes from applied, domain-specific innovation. This is where the UK can stand out — not just by advancing foundational models, but by championing platforms like Causaly that translate AI into measurable outcomes in high-stakes fields like life sciences R&D.