AI or no AI, that is the question! In a world where everyone seems to be talking about the technology’s tremendous potential, what is hype and what are the actual opportunities for your R&D and innovation teams. In a survey, we asked our digital community just that. Discover how many companies are already integrating AI, if there are sectorial differences and in which R&D phases are the most popular.
Who’s mastering AI and why
It’s amazing to see that most respondents are already using AI in their innovation process. So why are some companies already adopting AI and others not?
Companies integrating AI into their R&D processes likely see clear benefits, such as increased efficiency and improved product development. These companies often have a strong innovation culture, embracing new technologies quickly and encouraging experimentation. On the other hand, companies not yet using AI may face obstacles in setting up the necessary AI skillset and infrastructure. This is particularly challenging for start-ups and SMEs with limited budgets. They may lack in-house expertise, sufficient data, or the ability to manage data effectively. The return on investment for AI projects can be hard to quantify without prior experience, making it difficult to justify costs upfront.
AI adoption rate, total and per business type ©Verhaert
When comparing business sizes, start-ups are clearly taking the lead, but SMEs and especially large companies are making headway too. Why is this? Large corporations (49% using AI) often have the financial muscle to invest in AI expertise, infrastructure and experimentation. They also have vast amounts of data from existing operations, a valuable asset for AI. Initial costs can be a significant hurdle for SMEs (42% using AI), even though they are typically more agile and quick to adopt new technologies to remain competitive. Finding and retaining skilled AI talent can be more challenging for smaller companies, too. Many start-ups (85% using AI) operate on cloud-based infrastructure and have a data-centric culture, making it easier to integrate AI solutions. The start-up environment is often more attractive to AI talent, though early-stage start-ups may struggle with access to capital for AI investments.
Navigating value creation in new products and services
In addition to boosting the development process, AI can also add new and extra value to products and services. When looking at the results of the survey, there’s a clear difference in AI usage for this purpose between respondents who aren’t using AI yet in their development process (36%), and those who are (67%). This is not surprising as integrating AI into the development process is often easier and requires less investment thanks to free and easily accessible tools, which can then be a gateway to AI for new features.
Adoption rates in AI for value creation ©Verhaert
There are several possible reasons why respondents might be more likely to activate AI to create new added values through AI in products and services:
- Innovators may have more access to the resources required to implement AI in R&D.
- Companies that have already experimented with AI might be more aware of the potential benefits of AI in their products and services.
- These companies may have a more open and collaborative culture that is more receptive to new technologies like AI.
Exploring AI adoption across the industry spectrum
When you look at the adoption of AI between sectors, there are some big discrepancies. This may be because some sectors more easily see a high potential return on investment in AI and are thus more likely to adopt it. For example, finance and pharmaceuticals could gain significantly and directly from more personalized financial products or efficient drug discovery through AI. Their innovation and R&D processes are costly and complex so the impact on the efficiency of these processes is high. The manufacturing sector could be more cautious about investing in AI for R&D because the perceived benefits are less clear compared to other sectors.
Adoption rates of AI in R&D per industry ©Verhaert
Additionally, sectors like telecommunication generate or have vast amounts of data, making them well-positioned to leverage AI for R&D. For starters, this data could be used to train AI models and extract valuable insights. On the other hand, the construction sector might have a different volume of data compared to finance, limiting the ‘straightforward’ AI application in R&D.
Thirdly, sectors that have a strong embedded culture of innovation are more likely to embrace AI and implement it into their culture and processes. These sectors may have a more risk-tolerant
mindset and can readily allocate resources to explore benefits. The technology sector, for example, is known for being innovation-minded and may adopt AI in product and service development more quickly.
AI in action, examining adoption across R&D stages
From the survey results, we can derive how and where AI is currently already being used in the different phases of the R&D and innovation process.
Let’s start by clarifying those phases and the potential AI usage.
- In the ideation and exploration stage, new ideas are generated. AI can help brainstorm new product ideas, identify existing trends and analyze customer data to understand unmet needs.
- The next stage, feasibility assessment, evaluates the technical feasibility and market viability of a concept. AI can analyze data, assess the concept feasibility and viability and identify potential risks.
- During the concept development stage, promising ideas are turned into more detailed concepts. AI can create simulations, build demonstrators, develop prototypes, and conduct virtual testing.
- During the prototype testing and validation, a detailed version of the final product, physically or digitally, is created to test its functionality and gather user feedback. AI can automate some of the testing processes and analyze the data collected.
- The new product introduction stage focuses on launching the new product into the market. AI can optimize the product design and further investigate user research in the upscaling process.
Adoption rates of AI in R&D per development phase and business size ©Verhaert
But what does the survey say? Respondents who are already integrating AI in the R&D process, do so throughout different phases. Businesses seem to collectively prioritize integrating AI in the ideation and exploration phase, possibly because this is the most straightforward option thanks to tools like ChatGPT, Gemini and Copilot. Feasibility assessment and new product introduction receive comparatively less attention, while concept development, prototype testing and validation fall in between.
When comparing adoption within business types, start-ups appear to be more focused on feasibility assessment and prototype testing than others, possibly reflecting their need to validate ideas with limited resources. Large corporations seem to excel in ideation and new product introduction, leveraging their established resources and experience. SMEs end up in between, showing varying levels of activity across different stages.
Why AI integration varies across sectors
Having covered the differences between types of business, let’s now take a look at the differences between sectors. Some industries don’t yet use AI in all phases, explaining discrepancies with the average adoption rate.
Adoption rates of AI in R&D per development phase and industry ©Verhaert
AI adoption in concept development seems to be higher in machinery, construction, healthcare and pharmaceuticals. This is probably because AI can assist in analyzing vast amounts of data to identify patterns, develop new drugs, personalize services and predict customer churn. For machine builders, AI can optimize designs, making lighter, stronger, or more efficient machines. Additionally, through predictive maintenance, AI can analyze sensor data to predict potential failures and schedule maintenance proactively.
While almost every sector is investing in AI for ideation and exploration, sectors like consumer electronics, finance and economy and telecommunications seem to also emphasize the use of AI in prototyping testing and validation instead of ideation and exploration. There are several possible reasons for this. Prototyping, testing and validation are more scientific aspects of R&D than ideation and exploration. There are already established methods and tools, making it easier for companies to deploy AI in these phases. Prototyping and validation also reduce the risk and cost of failure by identifying issues early and ensuring new products meet customer needs.
Aerospace, manufacturing, healthcare, and pharmaceuticals highly adopt AI in feasibility assessment to manage financial risks, detect fraud, and define investment strategies. In aerospace and manufacturing, AI analyzes simulation data to assess the feasibility and safety of new concepts, mitigating risks.
When it comes to new product introduction, the sectors FMCG, machinery, and transportation and logistics show the highest adoption of AI. In these cases, AI can help design packaging, optimize settings to transition to the production phase or assist in finalizing the design for production upscaling. In addition, AI can help predict consumer demand and forecast the success of newly released products.
In conclusion
In conclusion, the integration of AI in R&D varies significantly across companies and sectors, driven by factors such as financial capacity, data availability, and innovation culture. Start-ups lead in AI adoption due to their agile, data-centric approaches, while large corporations leverage substantial resources and data. Across R&D stages, AI is already predominantly utilized in ideation and exploration, with varied emphasis in other phases depending on business type and sector. This diversity in AI adoption underscores the tailored strategies companies employ to harness AI’s potential.
