I am admittedly obsessed with Artificial Intelligence (AI), probably because, after decades of experience using technology across multiple industries—starting in telecommunications and most recently in healthcare technology—I’ve never seen anything move so fast. I designed my first patents in my 40s, and while I could have had three times as many if I had had the right mentoring earlier, I now have over 30 patents issued. Most of these are in the use of technology, particularly in healthcare—specifically Electronic Health Records (EHRs)—but not all. Many of them also cover areas like literary database management, chip design, security, encryption, blockchain, networking, social algorithms, automation, augmented reality, remote patient monitoring, predictive analytics, expert systems, and machine learning.
When I began my work, we didn’t use the term “Artificial Intelligence” in any of these patents. The concept was still too abstract, and we feared we’d be laughed at or misunderstood. It was more prudent to use terms that would now be considered part of AI, even though we didn’t label them that way at the time.
The phrase “most humans” in the title above reflects a broader point about AI: it is fundamentally mathematical and formulaic. Its roots trace back to databases. Let’s remember that software like Lotus 1-2-3 (released in 1983)—which emerged shortly after the IBM personal computer—gave users the ability to create formulas that automatically adjusted with any data changes. This revolutionized computing by making calculations interactive and intuitive, allowing anyone to see instant results whenever the data changed. These spreadsheet formulas laid the groundwork for the modern data-driven tools we use today.
Fast forward to today’s world, where data isn’t confined to one computer but is spread across millions of devices globally. People frequently update information simultaneously—consider, for example, multiple co-workers editing the same document at once. To solve the problem of merging these concurrent changes without causing conflicts or losing data, researchers created specialized database formulas known as Conflict-free Replicated Data Types (CRDTs). Unlike standard spreadsheet formulas, CRDTs use commutative operations, meaning the final result remains the same regardless of the order of changes. It’s like adding numbers: 3 + 5 is always 8, just as 5 + 3 is always 8.
Today’s AI systems rely heavily on data consistency, accuracy, and accessibility. Especially in large systems like ChatGPT, AI needs massive amounts of distributed data that are stored and updated continuously across multiple servers. AI uses parameters (ChatGPT, for example, has nearly 2 trillion of them)—internal variables within AI models that are adjusted during training. These parameters guide the AI model’s behavior and help it make accurate predictions and decisions based on the data it receives. Techniques similar to CRDTs ensure AI systems can manage data efficiently, update parameters smoothly, and provide reliable and consistent information to users worldwide. Just as Lotus 1-2-3 transformed individual data calculations decades ago, modern commutative database formulas now help power AI technologies, ensuring seamless interactions, robust learning, and trustworthy results. Historical innovations in data processing directly contribute to the capabilities and success of AI applications today.
AI’s Impact on Business Today
While the headlines from Washington D.C. may be grabbing all the attention, the real impact of AI is unfolding below the radar. Sure, there are major announcements from tech giants like Amazon and Oracle, investing billions in expanding Agentic AI systems to help businesses become more efficient, effective, and faster. But these developments also have significant implications for employment.
Someone once wrote, “AI is not going to take your job, but someone using AI may.” I am 100% convinced of this. While AI cannot perform many tasks today, the individuals and businesses that leverage AI effectively will be far more competitive. Lawyers using AI to draft court briefs or doctors relying on AI without caution may face serious risks. We’ve written about that before and will continue to explore the subject.
Just months ago, we were hearing about AI’s use in clinical care, but with many caveats. Now, we’re reading papers that show progress but also emphasize caution. AI can be helpful, but it requires significant time and effort from physicians to ensure its accuracy and reliability. Achieving a 95% accuracy rate from AI in healthcare will take time, and the question remains: Are we ready to accept AI’s 95% accuracy, particularly when it comes to our lives?
AI’s Impact: From the Top Down and the Bottom Up
A headline caught my attention recently: “PwC Cuts Record Number of UK Partners and Halts Tech Apprenticeship Scheme.” In a company that’s been around for 150 years, these changes are significant. PwC’s decision to cut 123 partners—double the historical average—along with suspending its technology apprenticeship program, signals a shift. Automation is clearly having an impact, with a focus on technology rather than human labor. We’ve seen similar reports from large accounting firms where profits have dropped.
I recently met with a major U.S. company, and their comments stunned me. They said, “We don’t need programmers.” Though they’re not in the tech industry, technology is incredibly important to them. They emphasized the need for just two types of employees: project managers and critical thinkers. It seems that PwC is adopting a similar mindset. I’ve mentioned in previous blogs that the first person I knew to be replaced by AI wasn’t directly replaced by a robot—but rather by fewer people handling the same work, thanks to AI. This happened in nursing and case management.
Health and AI: Historic Opportunities Ahead
In future blogs, we’ll delve into the fact that large language models (LLMs) are becoming a commodity. Many of them won’t generate significant revenue unless they pivot to Agentic AI and specific use cases. For example, there are now at least 20,000 AI agents in existence, and that number is growing rapidly. An expert I recently read about highlighted 150 potential AI use cases in healthcare. While he’s highly successful, I remain skeptical that all of these ideas will work as conceived.
What makes AI extremely dangerous in healthcare are the two polar risks: one in the back office and one in the front office. AI can be used for billing medical services, but this poses a real risk—fraud, overcoding, or overutilization—even with proper documentation. The U.S. government has already cracked down on technology companies in healthcare, and I predict that AI will soon be used in a way that makes major headlines—particularly in healthcare documentation and billing. Insurers and government contractors will rely more on AI to detect abnormalities, and we’ll likely see an “AI arms race” as a result.
On the other side, healthcare providers are seeking AI solutions that won’t slow them down or put patient care or malpractice risk at risk. Those are the guardrails AI will need in healthcare. While 100% of the AI companies in the aforementioned expert’s list were focused on payors, providers, and life sciences, I stand by my earlier point: in life sciences, AI has a critical role and will continue to expand as more data become available.
Consumer Data and AI
In the U.S., we have more health records on consumers than any other industry, and that volume is growing fast. As mentioned earlier, there are over 1.2 billion medical records, 3,000 exabytes of data across 500 EHRs. Imagine if we could mine this data longitudinally, perhaps augmented by each consumer’s genomic profile, to drive innovation in life sciences where AI could truly save lives. This would revolutionize the industry—without putting patients at risk or engaging in risky overbilling practices.
We have the capacity today, but until now, few have seen the consumer as the key player in this equation. In the 150 use cases I mentioned, none identified the consumer as the user. That’s about to change, and it will be a pivotal moment in healthcare’s AI revolution.
About HealthScoreAI ™
Healthcare is at a tipping point, and HealthScoreAI is positioning to revolutionize the industry by giving consumers control over their health data and unlocking its immense value. U.S. healthcare annual spending has exceeded $5 trillion with little improvement in outcomes. Despite advances, technology has failed to reduce costs or improve care. Meanwhile, 3,000 exabytes of consumer health data remain trapped in fragmented USA systems of 500 EHRs, leaving consumers and doctors without a complete picture of care.
HealthScoreAI seeks to provide a unique solution, acting as a data surrogate for consumers and offering an unbiased holistic view of their health. Giving Consumers tools to respond to denial of care by insurers, we aim to bridge gaps in healthcare access and outcomes. By monetizing de-identified data, HealthScoreAI seeks to share revenue with consumers, potentially creating a new $100 billion market value opportunity. With near-universal EHR adoption in the USA, and advances in technology, now is the perfect time to capitalize on the data available, practical use of AI and the empowering of consumers, in particular the 13,000 tech savvy baby boomers turning 65 every single day and entering the Medicare system for the first time. Our team, with deep healthcare and tech expertise, holds U.S. patents and a proven track record of scaling companies and leading them to IPO.
Noel J. Guillama-Alvarez
https://www.linkedin.com/in/nguillama/
+1-561-904-9477, Ext 355
https://www.ft.com/content/dee26dbe-8436-4f34-83eb-91cb18bf988d
https://www.ft.com/content/c1c1eaab-0eda-46da-b530-cf3d42c351eb