How does AI-enabled causal modelling support discussions with potential investors?

Welcome back to our blog series where we share some of the highlights and key takeaways from our fireside chat with CEO of Simbec-Orion, Fabrice Chartier, and CEO of, Joern Klinger.

Our last blog post covered what AI-enabled causal modelling is, how it is different from other models, and where you might want to start if considering adding AI to your clinical development plans.
This post continues the discussion on how early utilisation of AI could help de-risk and accelerate clinical development, by looking at how AI can demonstrate the potential of the pipeline to investors and how AI could disrupt clinical development as we know it today.

How can AI support smaller biotech companies at a time when investment is highly competitive?

Fabrice Chartier, CEO, Simbec-Orion
Any project or company requiring funding is currently facing significant challenges. For biotech companies with a limited number of assets, often only one, failing is not option.

We have seen investors pushing to get efficacy data early (often too early in my opinion), which could lead to risky protocols, or asking too much from the patients. Demanding protocols can lead to tired, demotivated patients or discourage them from joining, putting additional pressure on the trial. With the platform, we can provide critical information early, and de-risk the project without impacting the patient experience.

Is there an example of how causal modelling provides critical information earlier?

Fabrice Chartier, CEO, Simbec-Orion
We see the strategic partnership with as a clear way to provide earlier, more strategic support to our clients and to help them build a business case that is attractive to investors. Whilst demonstrating value to investors is just one part of the support AI can offer, this is just the ‘ignition’. Once investment is secured and development can move forward– you can utilise those insights to run a clinical trial that is more targeted and de-risked. As an example, some insights, such as measuring causal biomarkers, are usually gained in phase II. However, using’s platform, we can identify and measure the biomarkers in a FIH /phase I context, getting those insights and demonstrating causal relationships earlier.

With regulators such as the EMA[1] and FDA[2] driving the discussion of how AI can be utilised in drug development, how disruptive do you think AI will be to clinical development as we know it today? Is there cause for concern when there is also evidence of AI companies with great promise failing[3]?

Joern Klinger, CEO,
I have found drug development to be very conservative, but this is for a good reason. When it comes to using new computational technologies in clinical development, we should be skeptical, which is why we made the strategic decision to develop a technology that compliments the traditional clinical development process. When these AI companies with great promise, former ‘unicorns’ don’t live up to the expectations, it is because the expectations were misguided in the first place. Deep learning in particular can look impressive in other fields, like speech processing, and image recognition (which is how we have self-driving cars) and this generates a lot of ‘hype’ with investors and journalists – but there aren’t any grounds to assume that this would transfer to drug development.

The only area where this has translated well in drug development is latent discovery – designing novel proteins that can lock to a target site. Whilst there have been advances in latent discovery, it is too early to see its impact on drug discovery. Add to that the black box character of those models and you see why the impact has been modest and the hype is mostly gone. Whether AI will be used in the future – I hope so, but I think the path is to use technologies that can integrate well with existing workflows and the standards that have been proven for decades. We can develop these innovative, more speculative technologies as the industry develops new guidelines on how to incorporate new technologies into clinical development.

Fabrice Chartier, CEO, Simbec-Orion
I agree with Joern. We are in a very conservative environment, but that is because we are in a highly regulated industry. However, if changes are driven by the regulators, it will help promote these changes in a regulated environment. Disruptive, yes, but I think we are at a turning point. If the regulators are leading the discussion on using AI and machine learning in clinical development, changes will be implemented faster.

Disruption is not necessarily negative, disruption can accelerate progress. However, companies that do not consider the benefits of AI when their competitors are more open to leveraging innovative technologies may find themselves trying to catch up in future.

The failing AI companies in catchy headlines are not necessarily a sign that artificial intelligence or machine learning is going to fail. Being in the AI business does not protect you from going bust. Unfortunately, no AI tool can replace an effective management team. The fact that some AI businesses are going bust is a reflection that any business can go bust – it’s related to the business strategy. Perhaps the strategy was not right for their customers, perhaps it was too ambitious, or maybe it was not the right time. The failure of an AI business, even those ‘unicorns’ with great promise, is not an indication of the success or failure of AI or machine learning companies overall. In general, I believe AI businesses will be successful.

I do not believe that AI can replace what humans do – we have to see it as a new technology that can facilitate our understanding of a situation and support the decision-making process. With, we can provide reliable clinical predictions, which can help run clinical trials more efficiently – however, we still need trials to demonstrate if the predictions were correct.

If AI will lead to potentially fewer, smaller clinical trials – what does this mean for CROs?

Fabrice Chartier, CEO, Simbec-Orion
AI is a new technology, so we can try to predict what may happen, but we cannot determine its full impact at this stage. What I can confidently say, is that Simbec-Orion and other CROs are always trying to achieve success. Success for our clients, success for the drug – but most importantly, success for the patient. Whether AI will reduce the market or not, if we can increase the success of clinical development, it’s a good thing for everybody.

At the same time, whilst AI could lead to a reduction in unsuccessful early-stage clinical trials by utilising a more targeted approach – AI can also be used to indicate signs of promise for a different indication or patient group. As’s platform can identify other indications for certain drugs that have been abundant, this could lead to more clinical trials.

We cannot say for sure what the net impact on the overall market will be, but I believe we will be more targeted and more successful. Success usually drives investment, and investment usually drives activity. If there are more studies (which are more targeted and more likely to succeed), there will be more marketed drugs, which will benefit patients – so I’m very optimistic in that regard.

Joern Klinger, CEO,
Our partnership with Simbec-Orion would not have been possible without an open mind and consideration of the long-term impact of how AI could improve clinical development.
As a founder, I am also an idealist. Many times, we have had great ideas that were shut down because superficially it looked like they would reduce the market share for some potential customers, and I think that’s a short-sighted view. As Fabrice mentioned, with new technology comes new possibilities. It’s not sustainable to delay implementing innovative technologies because of the potential short-term impact, and again, we are predicting the potential short-term impact, as we cannot say for certain exactly how big an impact this will have on current processes and practices.

Our next blog will discuss how AI can help with orphan drug status, recent trends in clinical trial design and how AI can integrate with existing processes, and what data’s platform utilises to predict clinical outcomes.

[1]EMA (2023) Reflection paper on the use of artificial intelligence in the lifecycle of medicines, 19 July [online],for%20human%20medicines)%20or%20scientific
[2] FDA (2023) Discussion paper: Using artificial Intelligence & Machine Learning in the Development of Drugs and Biological Products
[3] O’Boyle, D (2023) The government said AI firm could ‘transform lives’ it was weeks from collapse, The Standard, 30 June

Back to Blog Archive