AI and Healthcare Innovation

Healthcare innovation must serve people first. Technology is the partner, not the master. That is the principle behind THE BRIDGE, and it is the thread running through my recent conversation in Prague with Lukas Benzel, Managing Director of the Czech Association of Artificial Intelligence, and my business partner, Peter M. Kovacs. Our focus was simple. What is real in AI for healthcare right now, what is next, and how do we move from headlines to outcomes?

Why AI in healthcare is different

AI is a general-purpose technology, but the health stakes are human. A better email subject line will not save a life. A better triage workflow might. That is why the measure of progress is not a demo. It is an earlier diagnosis, fewer complications, faster recovery, safer systems, and lower total cost of care. Evidence and adoption are non-negotiable.

Lukas shared a practical data point. More than 60 percent of Czech hospitals already use AI. Sometimes it is visible, for example, in radiology. Often it sits in the back office, scheduling, coding, or materials management. That mix matters. Fundamental transformation comes from a thousand minor improvements that free clinicians to practice medicine and free hospitals to operate with fewer bottlenecks.

Europe’s advantage, if we choose it

We do not have to win every AI race to be a leader. Building a general-purpose large language model that dominates the world is not the best use of Europe’s strengths. We can lead in applied AI for healthcare, defense, and the public sector. We have a strong base in clinical research, medical devices, and regulatory science. We can design systems that are safe, explainable, and aligned with public health goals. That is a value for citizens and a real market for investors.

Singapore offers a valuable reference point. Lukas described a cardiology center with an AI hub on the first floor that supports spinoffs, addresses operational challenges, and integrates research with care delivery. The lesson is not to copy and paste. The lesson is to align incentives and shorten the path from idea to bedside. When you put engineers, clinicians, regulatory experts, and investors at the same table, you reduce translation loss. You also reduce time to impact.

Prevention is power

We all know the line. We call it healthcare, but too often we run a sick care system. AI lets us flip that script. Continuous self-monitoring at low cost can move us from reactive to proactive. Think about what happens when wearables, home diagnostics, and clinical decision support work together. Risk signals surface earlier. People get nudges they will accept. Clinicians see the right patient at the right moment with the proper context. Insurers pay for fewer crises and more prevention. Employers keep talent healthy and engaged.

This is not an argument for gimmicks. It is a call to design for adherence and trust. A prevention tool is only helpful if people use it. The best way to earn that trust is to prove the benefit and make the experience humane. Clear privacy controls. Clear language. Clear next steps. And for higher-risk populations, a human touch point that does not disappear when the app closes.

Clinical trials and regulatory pathways

As we shift to prevention and personalization, clinical validation becomes even more critical. It is not enough to show that an algorithm performs well on a benchmark. We need to show that it improves decision-making and outcomes in a defined population under realistic conditions. This is where the right trial design matters. Adaptive trials and more brilliant site selection can shorten timelines and reduce cost without cutting corners. Real-world evidence can extend what we learn beyond the protocol. The destination is the same. Safer, faster, and more inclusive routes to market.

• For founders, here is a practical checklist I use when evaluating an AI health product:
  Clinical claim. What outcome do you improve, by how much, and for whom?

• Data quality. Do you have representative, consented data with a plan for bias detection and mitigation.

• Explainability. Can a clinician understand why the model suggested a path.

• Workflow fit. Where does this live in the clinical day, and who owns it.

• Regulatory path. Which class, which markets, which post-market commitments?

• Security posture. Zero trust by design, defined data retention, and vendor audits.

• Business model. Who pays and why, and what is the time to pay back for the buyer.

• Adoption plan. Training, change management, and measurable ROI milestones.

Hospitals, systems, and the back office revolution

It is tempting to focus solely on glamorous use cases, such as oncology imaging. Novel proteins. Personalized therapy selection. All important. But the near-term ROI in most health systems still comes from operations. When the admissions office schedules better, nurses spend more time with patients. When the pharmacy stocks smarter, fewer procedures are delayed. When the revenue cycle works the first time, administrators stop chasing paperwork and start solving problems. These wins are not dull. They are the fuel that keeps clinicians focused on care.

People, not just technology

Adoption lives or dies with trust. Older adults are often skeptical at first. That changes when a physician explains that AI is a second set of eyes, not a replacement. It changes when the benefit is concrete. A faster diagnosis, fewer false alarms, a safe discharge with real follow-up. Education is the strategy. Not a one-day workshop. A culture. Continuous learning for clinicians, managers, and patients. I ask teams to measure literacy the same way they measure uptime. If your team does not understand the tool, it will not deliver.

Small and medium companies also have work to do. AI is not a luxury reserved for large enterprises. Your accountant can use it. Your project manager can use it. Your quality team can use it. The trick is to find use cases with measurable return and to keep humans in the loop. Do not start by asking how many people you can cut. Start by asking how many good people you can keep by giving them superpowers. Train the team you have. Choose the right tools. Track the savings and the quality gains.

Cybersecurity and ethics are part of the product.

Safety is not a slide at the end of a pitch. It is part of the core design. Adversaries will use AI. That is already true. The correct response is not fear. The proper response is discipline. Data minimization. Encryption. Monitoring. Incident response that is practiced, not theoretical. Clear user education. And governance that fits the risk. If you are handling clinical data, act professionally and responsibly every day.

Europe is moving. So are we

National strategies and action plans are only valid when they reach the ward, the lab, and the startup. The priorities are straightforward. Implement AI in the public sector to improve services. Support basic research at universities and institutes. Help companies adopt AI responsibly with incentives tied to outcomes. The Czech ecosystem is ambitious. It is also realistic. We will not lead in everything. We will lead where we choose to focus and where we connect the right people.

What good looks like

I left Prague thinking about three pictures.

Picture one. A hospital floor where a triage model reduces wait times by thirty minutes, a pharmacy model cuts wastage by ten percent, and a scheduling model frees ten nurse hours a week. No magic.

Just fewer delays and better care.

Picture two. A clinical trial where adaptive design drops time to readout, underrepresented patients are recruited with the help of community partners, and real-world evidence supports a faster, safer launch. Investors get ROI. Patients get access.

Picture three. A prevention stack that people trust. A wearable that does not nag. A simple and accurate home test. A clinician dashboard that prioritizes the right alerts. And a reimbursement model that rewards staying healthy.

How to partner with me

My work sits at the intersection of innovation and adoption. I connect investors, innovators, and policymakers to turn evidence into action. I support founders who are ready for clinical validation and market access. I help investors who want deal flow that is curated for impact and ROI. I support hospitals and ministries that want to implement AI with safety and speed.

If this is your mission too, here are next steps:
• Founders. Please bring me your clinical claim, data plan, and adoption strategy.
• Investors. Let us co-invest in prevention-first tools and operational AI with proven ROI.
• Policymakers and payers. Pilot prevention with real incentives, then scale what works.
• Hospital leaders. Start with back office wins, then bring clinicians into the design loop.

The most powerful innovations are those that leave no one behind. Let’s connect to make prevention the standard, not the exception.

Timecode:

00:00 Welcome and Introduction

00:28 Overview of the Czech AI Association

01:09 The Importance of AI

02:18 Managing the AI Association

04:27 AI in Various Sectors

07:04 Global AI Competition

09:14 AI in Healthcare

10:21 AI in Singapore

18:19 National AI Strategy

21:42 AI in Everyday Life

22:37 The Future of AI

28:55 Cybersecurity and AI

30:30 Founding the AI Association

31:42 Joining the AI Association

Links:

Lukáš Benzl: https://www.linkedin.com/in/benzl/

Peter M. Kovacs: https://www.linkedin.com/in/petermkovacs/

Pavlina Walter: https://www.linkedin.com/in/pavlinawalter/

Website: PavlinaWalter.com

Episode Transcript:

Welcome everybody here in this nice location in Prague. Can you introduce us, because we have a new guest, a new member?

I’d like to introduce you. This is the head of the AI Association in the Czech Republic. It’s one of the most important associations now, as AI is the number one discussed topic. Please, introduce yourself and present the AI Association.

Thank you very much for the invitation. It’s a great pleasure for me. I am the managing director of the Czech Association of Artificial Intelligence. We are a typical NGO, but what is not typical is our focus on AI and all the sectors it will change in the future. We connect the dots between sectors, and right now we have more than 300 members. These members range from small AI startups to big tech companies like Google, Microsoft, or Meta.

Why is AI such a hot topic right now? It’s everywhere in the newspapers, and many countries, including the Czech Republic, want to become an AI hub.

It’s because AI is a real game changer for everything. It will change how we live, how we work, and how we spend our free time. We want to capture this opportunity for a better life in the Czech Republic. These are big words, but we truly believe it. AI is not only about technology, it’s about people and mindset. It brings positive impacts but also challenges. That’s why we talk not only about the technology, but also about the wider impact of AI.

How did you manage to set up this association? It’s a lot of work, with many divisions – companies, lawyers, education for kids. How do you manage it all?

We use AI. That’s the short answer. But the real answer is passion and hard work. We are also riding a huge wave of interest, because AI is such a big trend and hype. That hype is part of our success.

But how do you differentiate hype from meaningful opportunities? You have more than 300 members – how do you bring them to collaborate in ways that create real outputs?

That is the hard part. Everything starts with hype, but then you need business value and practical use cases. We start with discovery meetings, workshops, events – from small meetups to large conferences. A huge part of our work is education, showing companies what is meaningful and what is just trendy.

Which sectors are the most promising for AI in the Czech Republic right now?

Healthcare is number one. More than 60% of Czech hospitals use AI, sometimes in radiology, sometimes just in the back office. Education, defense, and the public sector are also very important. AI can help states be more productive and save money for citizens.

How strong is HealthTech in your association?

We have more than 20 companies in our healthcare working group. Some are established players developing software for hospitals, while others are startups using AI to detect tumors in radiology or cardiology. Some of them are very successful. Regulation is tough, especially with EU rules, but that is also a sign of quality.

How does Europe compare with the US and Asia in AI? Can Europe catch up?

Every continent wants to be a leader. Europe received a wake-up call from the US and China, and now we have a plan called the AI Action Plan. It includes building AI factories and giga factories – focusing on infrastructure, computing power, and independence. Right now, Europe relies on US technology, from chips to cloud systems, but we want to become more independent.

We cannot catch up in everything. For example, building large language models at global scale may not make sense. But Europe has strong areas – healthcare, defense, and manufacturing – where we can lead.

You just came back from Singapore. What did you see there?

Singapore impressed me. Their ecosystem works because they know how to connect players and spend money wisely. They already know their strengths, and healthcare is one of them. I visited a cardiology center with an AI hub solving important problems and supporting spinoffs. We have already imported ideas from Singapore, like the AI Readiness Index.

Could Czech hospitals realistically build AI hubs like in Singapore?

Not many, maybe one or two. The main limits are money, expertise, and digital readiness. AI transformation depends on digitalization and data systems, and we are still weak there.

In healthcare, diagnostics are already advanced with AI, especially in radiology. Where else could AI help?

AI can accelerate drug research and clinical trials – from selecting drug candidates to adaptive study designs. But many companies still focus only on diagnostics. I believe AI will increasingly focus on prediction and prevention – keeping people healthy instead of just treating illness. We need to move from “sick care” to true healthcare. Prevention is cheaper than treatment, and AI can make it affordable and effective.

How open are people to this? Are younger and older generations different?

Young people are naturally open. Older people are more skeptical at first, but when doctors explain that AI is not replacing them but supporting them, they become more accepting. In fact, many older people are excited about AI in healthcare once they see the benefits.

What is the government’s plan for AI in the Czech Republic?

We have a new national AI strategy with action plans under different ministries. The focus is on implementing AI in the public sector, supporting basic research, and helping companies adopt AI. The Ministry of Trade and Industry is driving this, and there is strong ambition at a national level.

Is the president supportive as well?

He has already met with leading AI researchers and seems to understand AI’s potential. He is modern and open-minded.

What examples can you give of AI in research and business?

AI is already creating new materials, new proteins, and potentially new drugs. It is also connecting massive datasets – from nature, from companies, from global sources – to generate new insights.

What about small and medium businesses?

AI can be used everywhere – in accounting, marketing, project management, IT. The challenge is finding the right use case with real return on investment. AI is for everyone, from freelancers to large enterprises.

Is AI just a bubble?

No. The level of investment is too high. New models and tools appear every day. AI is not going away; we are only at the beginning.

Some fear AI will take jobs. What is your view?

Four out of ten Czechs fear losing their jobs to AI. But the real issue is education and lifelong learning. If you stop learning, AI may replace you. But if you adapt, AI will empower you. Companies should not just fire people but upskill them to use AI tools.

How should kids be educated about AI?

Teach them that AI is a tool, not perfect, and that critical thinking and verification are essential. Encourage them to use AI for tasks they don’t enjoy, and develop their own skills in what they love.

Which AI tools do you use personally?

ChatGPT is my daily tool. I also use Brain.fm, AI-generated music to boost focus, and other AI platforms for specific tasks.

What about security and ethics?

Be cautious: don’t share personal data, and read the fine print of AI tools. At the same time, AI will be essential for future cybersecurity – it’s the only way to fight against malicious AI systems.

How did the idea of creating the association come up?

I have always loved technology. When generative AI became available for everyone – not just tech experts – I realized it was the right time. I did my research and saw no one in the Czech Republic was addressing AI in such a comprehensive way, so we founded the association.

How can people join?

It’s simple. Get in touch, we’ll meet online or in person, discuss how we can help, and then you just fill a short form on our website.

Thank you so much.

Thank you.