AI in healthcare: From hype to human-centred, scalable solutions

The future of AI in healthcare depends on thoughtful adoption: solutions shaped by need, aligned with UK standards and designed to enhance, not displace, human expertise

The recent Forbes article, AI And Hybrid Healthcare Models Are Driving The Next Wave Of Scale by Gary Drenik, offers a compelling narrative about how artificial intelligence (AI) and hybrid care models are beginning to reshape the clinical experience. The article opens with a simple but powerful scene: a clinician who can sit, focus, and speak directly with a patient without being buried in systems, screens, or fragmented workflows. In a sector where administrative burden has steadily eroded the time and attention clinicians can give to the people they serve, this vision is both refreshing and necessary.

Yet, as promising as these developments appear, they also highlight deeper tensions in the global health and care landscape. The rapid acceleration of AI solutions, driven by hype cycles, competitive pressure and industry enthusiasm, risks repeating longstanding mistakes. Too many tools are being built first, with problems identified second. Too often, innovation is shaped by what technology can do, rather than what frontline teams need it to do.

As organisations across the NHS, independent healthcare, adult social care, and wider health services explore the use of AI, predictive analytics and hybrid models, the challenge is not simply to adopt new systems, but to do so safely, collaboratively and with a clear understanding of the operational realities facing providers.

This article explores the key themes raised by Drenik’s piece, examines what they mean for the UK context, and reflects on how health and care leaders can harness AI meaningfully, not as a panacea, but as a thoughtfully deployed enabler.

Drenik highlights a truth recognised by clinicians worldwide: digital transformation has not yet delivered the simplicity it promised. Instead, it has frequently produced fragmented systems, labour-intensive documentation and a growing sense of digital fatigue. This is evident in the UK as much as in the US and globally.

NHS, independent health and social care providers grapple daily with:

• Duplicated data entry
• Disjointed electronic record systems
• Poor interoperability
• Reporting obligations that consume frontline time
• Legacy systems that fail to support real-world workflows.

These challenges undermine the very purpose of digitisation: to free up clinicians to spend more time with patients.

AI-enabled workflows, such as virtual scribing, automated scheduling and intelligent templates, are now emerging as powerful tools to relieve this burden. As Drenik notes, organisations like Gameday Men’s Health are already using AI-supported EMR systems to reduce documentation time from hours to minutes. These examples reinforce a critical point: AI provides the greatest value when it removes invisible friction rather than adding new tasks.

However, this requires the right design philosophy. Tools must be intuitive, embedded within existing clinical processes and aligned to regulatory requirements. They also need rigorous governance to ensure safe and compliant deployment.

One of the most important reflections from the UK perspective is the pace at which AI tools are being pushed into the health and care ecosystem. The demand for technological solutions is real, driven by workforce shortages, rising demand, and persistent operational inefficiencies. But the rush to deploy AI also carries risks.

There is a growing trend for vendors and developers to build solutions powered by the latest AI capabilities, then search for problems to attach them to. This technology-led approach rarely lands well in complex, highly regulated environments such as health and care settings.

There is a growing trend for vendors and developers to build solutions powered by the latest AI capabilities, then search for problems to attach them to. This technology-led approach rarely lands well in complex, highly regulated environments such as health and care settings.

Safe and successful innovation requires:

• Deep engagement with frontline teams
• Understanding the realities of different clinical settings
• Co-design with providers, practitioners and service users
• Alignment with UK regulatory, ethical and data protection standards
• Continuous evaluation and improvement.

Without this, AI risks becoming another layer of complexity, another system for clinicians to navigate, another interface competing for attention, another source of administrative noise.

The lesson is clear: technology should follow need, not the other way around.

Drenik’s article reinforces a critical insight: despite rapid technological progress, the vast majority of patients still prefer to communicate with human clinicians. In the referenced Prosper Insights & Analytics survey, just 5.6% of US adults would be willing to use AI alone for healthcare matters, and nearly 86% preferred interacting with a person rather than an AI agent.

Although the data reflect a US population, the sentiment closely aligns with broader international findings. In the UK, trust in health and care is built on continuity, relationships, and the human qualities of compassion, judgement and empathy. AI may support decision-making and streamline workflows, but it cannot replicate the reassurance a patient feels when speaking with a knowledgeable clinician who understands their history and concerns.

This is why any meaningful AI adoption must preserve, and ideally enhance, the clinician–patient relationship.

This is also where hybrid models show immense promise.

The old debate between digital and face-to-face care is fading. Drenik highlights how hybrid models are becoming the dominant model in the US, and the same pattern is emerging rapidly across UK health and care services.

Hybrid models combine:

• The convenience of digital access
• The safety and relational depth of in-person care
• Real-time data integration
• Continuity with the same clinician or care team.

This approach has clear benefits:

• Improved patient adherence
• Higher retention
• Better clinical decision-making
• Greater accessibility for people with work, family or mobility challenges
• More efficient use of clinical time.

In the UK context, hybrid models also support broader NHS ambitions around digital access, integrated care pathways and improved population health outcomes.

But hybrid care is only effective when the digital and physical components are joined up. Fragmentation, the very issue Drenik’s article repeatedly points to, remains the most significant barrier.

A particularly strong section of Drenik’s article focuses on predictive analytics and operational forecasting. Drawing on models used in sectors like logistics and aviation, predictive tools can anticipate no-shows, smooth appointment flow, optimise staffing and guide supply decisions.

This shift from reactive to proactive management is essential for scaling services safely.

In the UK, similar models have already shown promise within urgent care, community health and social care scheduling. By anticipating demand patterns, providers can allocate resources more effectively and reduce pressure on overstretched teams.

However, predictive systems must be transparent, explainable and tightly governed. Unlike the commercial sector, health and care settings carry far higher stakes: decisions informed by predictive models can directly affect safety, equity, and access. Robust governance frameworks, ethical safeguards and regulatory compliance must be built into every stage of design and deployment.

Where Drenik’s article provides a high-level perspective on emerging models, the UK context adds an additional layer of complexity. Any AI system deployed within health or social care must align with:

• NHS England guidance
• CQC and Ofsted regulatory expectations
• Data protection and information governance legislation
• NICE evidence standards
• Clinical safety standards, including DCB0129 and DCB0160
• Workforce training and competency requirements
• Ethical frameworks set by relevant professional bodies.

Rapid adoption, without adherence to these standards, can pose significant risks, from data breaches to unsafe clinical decisions.

The current wave of AI enthusiasm is creating pressure to implement solutions quickly, often without sufficient understanding of the operational realities those tools must support. A safer and more sustainable approach is to co-design AI and digital systems with providers, educators and practitioners themselves.

This ensures the tools address real problems, work in real settings, and support regulatory compliance from day one.

Across the global health and care landscape, a common theme emerges: the most transformative AI is the kind that disappears into the background. It enhances safety, efficiency and continuity without interrupting the clinical relationship.

Human-centred AI:

• Supports clinicians rather than competing with them
• Strengthens trust rather than weakening it
• Provides clarity rather than complexity
• Integrates with existing workflows rather than disrupting them
• Improves consistency across teams, sites and services.

This is the real future of AI in health and care: not replacement, but restoration, giving clinicians the time, space and information they need to care well.