Pharmaceutical Market Europe • October 2023 • 14

MIKE DIXON

MIKE DIXON
DID SOMEBODY SAY AI?

Things are moving rapidly and we are currently only scratching the surface of AI’s possibilities

Image
Image

Despite being around and indeed utilised in pharma for several years, 2023 has seen the world fully wake up to the potentials and challenges of AI. I think it is probably fair to say that ChatGPT, its subsequent iterations and then big tech’s rapid integration of AI into search engines have been significant catalysts for AI becoming the hot topic for almost everything. Let’s face it, AI comes up everywhere; mainstream media, on social and at forums, conferences and events. I am also sure most of us by now have also had a ‘play’ with AI in some form or other. And therein lies the first potential challenge.

What are people trying and have they considered what they are doing and the potential consequences? No, I am not talking Terminator-level destruction of mankind, but rather more inadvertent copyright infringement or confidentiality breaches. Off the record, I have heard various examples where an individual has put commercially sensitive information into a publicly available AI like ChatGPT, without considering the risk in doing so. Unreleased trial data, sales and financial data, HR information, the list goes on. Although there will be debate among security experts on the level of risk for different scenarios, there is no doubt that informed policies and processes should exist within organisations to ensure the responsible use of generative AI.

Against this background, the Healthcare Communications Association (HCA) has published their road map charting how we should approach AI in healthcare communications. It is available at Current Medical Research and Opinion (ICMO):
https://doi.org/10.1080/03007995.2023.2261737 or via the HCA website.

Waiting is not an option

We certainty can’t afford to just wait and see what happens with AI before deciding what we do. For the reasons already mentioned, AI is here and accessible to everybody, so we need to ensure any use is responsible, appropriate and ethical at every level. To achieve that we need to decide how we are going to approach AI and put key principles and policies in place to communicate our intentions. This isn’t about alarming people, but just ensuring there are clear guard rails in place to allow innovation to continue in a way that protects individuals, their organisations and clients. The road map also points out that taking a blanket ‘no AI’ attitude is probably not a viable option, as it will potentially leave you behind the innovation curve, while also risking shadow use occurring outside the security, regulatory and legal frameworks.

The first thing that I often hear and that I think articulates the situation well is that AI may not necessarily take your job, but if you don’t embrace AI within your job then you will lose out to those who do. There is no doubt that AI has the potential to provide valuable help in many aspects of our work in healthcare communications and marketing. There are probably some obvious areas to start exploring when clear policies and guard rails are in place.

Areas to explore

Learning – there is no doubt that AI can help us identify information and keep up to date in our ever changing working environment and the overwhelming amount of information potentially available. However, in using AI we must not lose sight of some of the instinctive questioning we exhibit ourselves. What is the source, is it credible and reliable, is it current? Just because AI finds it and uses it does not mean we should. Of course, AI may also be used as a tool to root-out fake or inaccurate information.

Writing – remembering the cautions above, AI can provide a great way to rapidly produce written summaries of information. What is produced should always be considered a first draft and be reviewed by a subject matter expert to ensure accuracy.  But the potential timesaving advantages for pitches or getting team members briefed are very appealing.

Creativity – I can imagine some creatives being apoplectic when discussing AI for creativity, but we should be open to the idea of AI being used at appropriate points in the creative process, whether to provide stimulus or for delivering simple layout ideas for slides, as an example. The next generation of AI tools reaching us will offer more opportunities with images and video, so this discussion and use is only going to increase.

Patient insight gathering – potential for rapid and comprehensive insight gathering on patient experiences is available using AI. These can help shape interactions or communications to ensure they better reflect real-world patient lives and experiences.

Clinical trial efficiencies – first drafts of required materials, such as patient lay summaries, are now being created by generative AI in some organisations. The process also has the ability to pitch content at the right literacy level for the audience being targeted.

The questions to ask

Ultimately, in considering where best to incorporate AI, the questions should focus on identifying where there are organisational challenges and whether AI can help in addressing these, such as time to production, quality, engagement/enjoyment and efficiency.

Metaphorically, the AI genie is now out of the bottle. How we use it is the question we all now face. Things will move rapidly, and we are currently only scratching the surface of the possibilities, so it is important we don’t get left behind. But setting out the right plan for the journey before you set off is vital to ensure you don’t get lost, or worse fall off a cliff, along the way.


Mike Dixon is CEO of the Healthcare Communications Association (HCA) and a communications consultant