Why won't this expert's clients sign onto AI projects for more than 12 months at a time?

Andy Sajous, a leader in digital transformation, explains. He discusses the trade-offs of building versus buying artificial intelligence tools and describes some crucial actions CIOs should take going into 2025.
By Bill Siwicki
01:08 PM

Andy Sajous, field CTO and healthcare practice lead at digital transformation firm Ahead

Photo: Ahead

The AI hype cycle is in an unusual place right now, especially looking at the court of public opinion versus how the professional sector portrays the technology's potential.

Andy Sajous is field CTO and healthcare practice lead at digital transformation firm Ahead. In all of his meetings with healthcare chief information officers and other IT decision-makers, none of them are willing, he says, to sign on to any specific AI product or service for longer than 12 months.

In this interview, Sajous explains why he thinks that is. He describes some of the fast-moving shifts in the AI market, the challenges of build vs. buy – he describes some key actions healthcare CIOs should be taking as we head into a 2025 that's poised to see even more AI transformation.

Q. You say in your digital transformation work this year with CIOs and other tech decision makers in healthcare, none of them will sign on to any AI product or service for longer than 12 months. What do you take away from these experiences and what do you think that means for AI's future in healthcare?

A. The reluctance to commit to AI contracts longer than 12 months reflects a profound uncertainty in the AI landscape within healthcare. CIOs and other decision-makers are wary of overcommitting to tools in an environment that's rapidly evolving.

AI vendors are constantly releasing new products, but the market is flooded with startups and smaller companies whose futures are uncertain. There's a real concern that a system that seems promising today could become obsolete within a year, or worse, the company behind it could be acquired or go out of business entirely.

The rate of change in AI technology, especially after the launch of generative AI tools like ChatGPT, has created a climate where healthcare organizations are forced to think short-term when adopting new technologies.

However, this does not indicate a complete lack of belief in the potential of AI. On the contrary, healthcare organizations are acutely aware of AI's potential to transform patient care, improve operational efficiency and streamline administrative processes.

However, they also recognize the technology still is in a state of flux, with new players entering and exiting the market constantly. CIOs are looking for flexibility, and that means being able to pivot quickly if a better technology emerges or if an AI tool they've invested in fails to deliver the expected results. They want to avoid being locked into long-term contracts with vendors whose products may not keep pace with the rapidly advancing state of the art.

For AI's future in healthcare, this cautious approach may slow down adoption in the short term but could ultimately drive a more thoughtful and strategic integration of AI into healthcare workflows. As the market matures and more stable, proven systems emerge, we may see healthcare organizations become more comfortable with longer-term commitments.

Until then, flexibility and adaptability will remain key. The healthcare sector will need to remain agile, continuously evaluating new technologies while ensuring that patient care remains uncompromised by unproven or rapidly outdated systems.

Q. You cite a rapid shift in market leaders in AI. Who have been and currently are the market leaders, and why the shifts?

A. The dynamic nature of AI means today's market leaders may not be tomorrow's. The AI landscape has seen significant shifts in terms of market leadership due to both innovation and consolidation. A few years ago, major tech companies like IBM Watson and Google's DeepMind were pioneers in healthcare AI, particularly in areas such as diagnostic imaging and predictive analytics.

With rapid development and new AI players, the market has continued to expand. Startups and niche companies are emerging with highly specialized systems, catering to very specific healthcare needs, such as AI-driven clinical decision support or AI-based diagnostic tools for radiology and oncology.

Companies like NVIDIA, which provides the hardware backbone for AI development, have become indispensable, especially in areas like machine learning and computer vision. Epic, which integrates AI into its electronic health record system, also is making significant inroads by offering comprehensive, AI-augmented systems more tightly integrated with existing hospital workflows.

These companies are leveraging their broader platforms to introduce AI capabilities, which could make it harder for smaller, more specialized vendors to compete unless they offer a truly unique value proposition.

The shifts in market leadership are driven by several factors. First, the rapid pace of AI innovation means that vendors need to constantly update and improve their offerings to remain competitive. Second, the consolidation of AI technologies into larger platforms, like Epic, reduces the need for standalone AI vendors.

Lastly, many healthcare organizations are still navigating the regulatory and ethical concerns around AI, which means that companies that can provide not just innovative systems but also trusted, secure and compliant systems will ultimately lead the market. These shifts indicate the AI landscape will continue to be volatile until a few clear leaders emerge.

Q. What are the challenges of building versus buying AI tools in healthcare?

A. The decision to build or buy AI tools in healthcare is not straightforward, and each path presents its own set of challenges. Building AI tools internally allows healthcare organizations to tailor the systems specifically to their needs. They can develop models aligned with their unique data sets and workflows, ensuring AI systems are finely tuned to the demands of their organization.

However, this approach requires significant resources, both in terms of financial investment and technical talent. Many healthcare organizations face a shortage of skilled AI professionals, and the costs of hiring and retaining such talent can be prohibitive. The ongoing maintenance and updates required to keep internally developed AI tools current with the latest advancements in the field can place a further strain on resources.

On the flip side, buying pre-built AI tools offers a faster route to implementation, with less upfront development effort. These tools often come with vendor support, which can help healthcare organizations get up and running quickly.

However, this approach is not without risk. The healthcare AI market is crowded with vendors, many of which are startups that may not be around for the long term. CIOs have expressed concerns about committing to vendors whose products may not evolve at the same pace as the needs of the organization or whose business models may not be sustainable.

Additionally, pre-built AI tools may not integrate seamlessly with existing health IT, leading to inefficiencies and potentially hindering the effectiveness of the technology.

Another key challenge when buying AI tools is vendor lock-in. When a healthcare organization becomes reliant on a particular AI tool, it can be difficult to switch to a different tool down the line if the vendor stops innovating or if a better system becomes available.

This can lead to a situation where the organization is stuck with a suboptimal tool, or worse, where the vendor goes out of business and leaves the health system scrambling for alternatives. Healthcare organizations need to carefully weigh the risks and benefits of building versus buying AI tools, considering not just the immediate costs and benefits but also the long-term implications for their IT infrastructure and patient care.

Q. What are key actions healthcare CIOs and other health IT leaders need to take going into 2025?

A. As healthcare organizations look toward 2025, CIOs and health IT leaders must focus on three critical areas: cloud optimization, talent development and data governance. Cloud optimization is crucial because many healthcare organizations are operating in a hybrid cloud environment, with both on premise and cloud-based systems.

Optimizing cloud usage not only enables scalability and flexibility but also helps reduce costs – an increasingly important factor given the financial pressures many healthcare organizations face. Ensuring their cloud infrastructure is both secure and efficient will allow health systems to leverage AI and other emerging technologies without being bogged down by legacy systems or exorbitant infrastructure costs.

Talent development is another key area where CIOs need to concentrate their efforts. There is a significant talent gap across the tech industry, but especially in health IT, particularly when it comes to AI and cloud engineering. CIOs must invest in training programs to upskill their existing staff, while also finding creative ways to attract new talent in a highly competitive market.

This might involve forming partnerships with educational institutions, offering specialized certification programs or working with vendors to provide joint training initiatives. Upskilling internal teams will be critical to ensuring that healthcare organizations can not only implement cutting-edge technologies but also maintain and evolve them as the industry continues to advance.

Finally, data governance is a top priority for healthcare leaders as they head into 2025. As AI and data analytics become more integrated into healthcare operations, ensuring the security, privacy and ethical use of patient data will be paramount. This involves implementing strong governance frameworks that can manage the vast amounts of data being generated, while also complying with regulatory requirements like HIPAA.

Moreover, CIOs need to be proactive in developing strategies to address potential risks associated with AI, such as bias in algorithms or data privacy concerns. Building a strong data governance infrastructure will be critical not only for mitigating risks but also for fostering trust in AI-driven healthcare tools.

Follow Bill's HIT coverage on LinkedIn: Bill Siwicki
Email him: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication

Want to get more stories like this one? Get daily news updates from Healthcare IT News.
Your subscription has been saved.
Something went wrong. Please try again.