Hey everyone! AI change is coming for every industry, but this week we are zooming in on healthcare. When AI moves into high-stakes parts of life, it stops being a novelty and starts forcing real shifts in power, expectations, and trust.

🏥 Founder’s Corner: Healthcare is the focus this week. I break down what OpenAI’s ChatGPT Health launch signals for the future and why I believe we are entering the era of the “Executive Patient.”

🧠 AI Education: ChatGPT Part 2 covers the “create and understand” layer so you know when to use Search, Deep Research, voice, and images to get consistently better results on the free tier.

🧾 10-Minute Win: A Tax Deduction Finder that helps you generate a realistic shortlist using IRS sources, then build a simple tracker so you are not scrambling at tax time.

1) Introducing ChatGPT Health

Summary: OpenAI launched ChatGPT Health, a dedicated health/wellness experience where you can (optionally) connect medical records and wellness apps to ground responses in your own data. It’s positioned as support for understanding and planning—not diagnosis or treatment.

Why it matters: This is AI moving into high-stakes daily life. If it works well, it becomes a “default tool” category for millions—while raising the bar on privacy, trust, and what people expect from AI assistants.

2) Samsung to double mobile devices powered by Google’s Gemini to 800m units this year

Summary: Samsung says it plans to expand Gemini-powered AI features from roughly 400 million devices to 800 million in 2026 across phones/tablets—and broadly push AI across products and services.

Why it matters: This is what “AI going mainstream” actually looks like: distribution at device scale. When AI ships by default to hundreds of millions of people, usage shifts from novelty to habit.

3) Global AI Adoption in 2025—A Widening Digital Divide

Summary: Microsoft’s AI Economy Institute reports global generative AI use rose in H2 2025 to roughly one in six people, but adoption is uneven—growing faster in the “Global North” than the “Global South.”

Why it matters: The next phase of AI isn’t just “who has the best model”—it’s who gets access, skills, and infrastructure. That divide is going to shape productivity, education, and competitiveness.

4) NVIDIA Rubin Platform, Open Models, Autonomous Driving

Summary: NVIDIA used CES to spotlight its Rubin platform roadmap and a broader “AI factory” vision: faster training/inference economics, more efficient systems, and new model + autonomy pushes tied to real products.

Why it matters: This is NVIDIA telling the market: the next wave isn’t “more GPUs,” it’s rack-scale platforms and efficiency. That shapes cloud pricing, enterprise adoption, and the economics of every major model provider.

5) xAI Raises $20B Series E

Summary: xAI announced a $20B Series E round and framed it as fuel for aggressive compute expansion and product rollout. It highlights massive GPU cluster ambitions and says Grok 5 is currently in training.

Why it matters: This is the clearest signal yet that the frontier AI fight is becoming a capital + compute arms race—and that the market expects only a few players to afford the next leap.

The Executive Patient: How OpenAI’s Health Launch is Rebuilding the Healthcare Experience

On January 7, 2026, OpenAI launched its new health platform within ChatGPT. This is more than a new feature for a tech company. It is a collision of two massive trends that the traditional healthcare system is not prepared to handle.

I have spent the last 13 years in the healthcare industry working across various roles to improve the experience for healthcare consumers. Currently, I operate in the digital space to deliver experiences and platforms that bring convenience to people dealing with specialty conditions. I have seen firsthand how hard it is for the system to catch up to the consumer's needs.

The reality is that healthcare consumers have been demanding change for a long time. The last decade has seen a massive rise in health tech, ranging from enterprise-scale patient apps to niche platforms focused on specific wellness needs. Consumers have shown their preference for these tools by adopting isolated digital platforms like Hims & Hers, Function Health, and Peloton. With OpenAI reporting that 230 million people already ask ChatGPT health or wellness questions each week, including 40 million daily queries in the U.S. alone, the launch of ChatGPT Health is not a new trend. It is the consolidation of a behavior that is already fully formed.

Over the last decade, we have also seen massive consolidation across every sector of the healthcare industry as companies try to solve fragmentation through mergers and acquisitions. Yet, the struggle to deliver a cohesive digital experience remains. It is incredibly difficult to build a seamless platform when you are constantly integrating new systems, defining legacy data, and shifting business priorities. While the industry works through these structural hurdles, consumers are already moving toward tools that offer immediate utility and a unified experience.

OpenAI’s partnership with b.well provides the underlying infrastructure that traditional health systems have struggled to build internally. This foundation effectively turns ChatGPT into the centralized health hub that consumers have been waiting for. However, my point is not to crown OpenAI as the definitive winner in this space. Instead, I am highlighting a shift where the winners of the future will be those who stop trying to own the entire journey and start figuring out how to deliver value within these AI-driven ecosystems.

We are witnessing the rise of the "Executive Patient," and the system is not ready for it. Here are the four transitions that will define this new era and my predictions for how the industry must adapt.

Pillar 1: The End of the Patient Portal

For years, the patient portal was the industry's attempt to go digital. Instead, it created a mess of fragmented apps. Every health insurer, hospital system, pharmacy, and specialist created their own proprietary data silos to protect their information and their relationship with the patient. For a person with complex needs, access meant managing a dozen different logins and trying to piece together a story from multiple disconnected interfaces. It was a fragmented experience that prioritized the institution's silos over the patient's experience. This is exactly what b.well CEO Kristen Valdes described as "portalitis" during the launch. She noted that we have spent years expecting consumers to do the heavy lifting of data integration themselves.

The release of ChatGPT Health changes that dynamic. By letting consumers sync medical records and wellness stats into a single experience, OpenAI has built a universal translator for healthcare. This creates a level of data liquidity that makes the traditional proprietary portal feel like a relic of the past. This integrated experience, which puts data directly in the healthcare consumer’s hands, will be the death of siloed apps across the healthcare ecosystem.

Over the next 24 months, consumers will stop logging into proprietary portals to find answers. Instead, they will use AI hubs like ChatGPT Health as the interface to connect all of their data and wellness tools into one experience. Power dynamics will force change at a rapid pace, and while this speed is unfamiliar to the healthcare industry, it is now unavoidable.

The winners of the future will be the companies that abandon the closed-loop ecosystem model. The future belongs to entities that make their data easy to interact with and focus on being the best service provider within the AI ecosystem, rather than trying to build a fragmented ecosystem of their own.

Pillar 2: The Rise of Patient Agency

The traditional healthcare model is built on information asymmetry. The doctor holds the clinical data, the medical training, and the authority, while the patient holds the symptoms and a hope for a solution. This dynamic is rooted in a fragmented reality where your health data is scattered across primary care visits, claims, and specialist referrals. Because these entry points rarely talk to each other, the practitioner only sees a narrow, clinical slice of the story.

The gap between these clinical records and a person’s actual life is where health is truly won or lost. I have experienced this disconnect in my own health journey. While the legacy system relies on a yearly snapshot, I am generating a constant stream of high-fidelity data that my doctor cannot access. My Oura ring captures my physiological stress and sleep cycles in real time. My comprehensive blood work through Function Health provides a longitudinal look at my biology that goes far deeper than a standard blood draw through my insurance. My nutritional habits and metabolic shifts are audited through tools like Noom and MyFitnessPal. In the traditional system, these insights are invisible. They exist in a separate world from the medical record, leaving the practitioner blind to the variables that drive my health.

The release of ChatGPT Health changes this dynamic by acting as the bridge between the lifestyle stream and the clinical snapshot. When a patient takes these personal insights and blends them with their medical history through an AI hub, the paradigm shifts. They are no longer arriving at the clinic to ask what is wrong. They are arriving with a 360-degree audit of their own biology to discuss what needs to change.

This shift is moving faster than the industry realizes. A January 2026 survey from The Mesothelioma Center found that 52% of Americans are now using ChatGPT to analyze symptoms, and one in three would skip a doctor visit entirely if the AI characterized their risk as low. We are also seeing patients like Jennifer from Wisconsin, who recently shared with the Advisory Board how she uses AI to synthesize medical literature. She uses these insights to confront physicians who previously brushed off her concerns.

This is no longer just about access to information, it is about the redistribution of agency. We are entering an era where the machine is more capable of synthesizing these massive datasets than any human brain. The patient is becoming the strategist of their own health journey, using AI as the engine to manage complexity. By the time they sit down with a specialist, they are not just a subject of the treatment plan. They are an active partner in the strategy, and the traditional hierarchy of the exam room has been replaced by a model of shared accountability.

Pillar 3: The Evolution of the Practitioner

Despite the rapid surge in AI capabilities, we are not at a point where technology can replace the clinician. Professional guidance remains the bedrock of a safe and effective health journey, and ChatGPT Health is an assistant to that process rather than a substitute for medical expertise. However, a patient with real agency and more information will force change across the entire healthcare industry. Practitioners will have to adapt and evolve if they want to keep and earn the trust of the executive patient.

The biggest hurdle in this transition is human nature. Change is difficult, and it is impossible for all doctors to adapt at the same pace as the AI space. This uneven adoption is already creating significant friction. A January 2026 report from The Mesothelioma Center highlights this tension, revealing that 58% of healthcare professionals now say AI is making it harder to treat patients. Dr. Daniel Landau noted in MobiHealthNews that clinicians often find themselves on the defensive, needing to disprove AI-generated findings rather than focusing on a collaborative diagnosis. This shift can turn medical visits into debates, which slows down appointments and risks damaging the patient-provider relationship.

This friction exists because the practitioner’s monopoly on interpretation is over. We are entering a state of clinical symmetricality where the patient may know as much about their specific health metrics as the doctor does. The clinician must shift from being "The Oracle" to becoming "The Strategic Partner."

The good news is that the tools are evolving to support this shift. Athenahealth’s CMO, Dr. Nele Jessel, recently stated, "2026 is the year the EHR finally learns to think. We are moving beyond the era of the digital filing cabinet and into the era of clinical intelligence, where the software doesn't just store data, it acts as a cognitive partner to the clinician." By moving from a passive system of record to a real-time partner in intelligence, technology can begin to shoulder the cognitive load of data synthesis.

The value of a doctor is shifting from being a primary data source to becoming a high-level advisor who operates in a collaborative environment. Trust is no longer a default setting based on a medical degree, it is something earned by practitioners who use AI to drive the best possible health outcomes. Doctors who remain anchored in the past and refuse to adapt will struggle in the healthcare model that is evolving in front of our eyes.

Pillar 4: The Living Health Record

The final pillar of this transformation is a fundamental change in how we store and interact with our medical history. For decades, the medical record has been treated like a static file locked in a high-security vault. It was something owned by the hospital or the insurer, and accessing it required the patient to navigate a labyrinth of requests and approvals. The launch of ChatGPT Health represents a "Napster moment" for the healthcare industry. By using b.well as a technical backbone, OpenAI has utilized FHIR APIs to bypass the need for individual hospitals to approve data sharing. The consumer simply hits "connect," and the walls of the silo finally crumble.

This transition turns our medical history into a "chatable" entity. Your health data is no longer a collection of static PDFs or disconnected laboratory results, but a living record that stays with you. OpenAI has designed its Health Sidebar to be isolated and encrypted, which creates a permanent history that is not tied to a specific insurance plan or a single employer. This is a fundamental shift in ownership. Data is no longer a static piece of information that belongs to a facility, it is a living entity that travels with the human rather than staying with the provider.

The implications of a truly portable health record are massive. When your history is no longer trapped in a specific system, the friction of changing doctors or switching insurance plans disappears. We are moving toward a future where the consumer is the primary custodian of their clinical truth. Because this record is "chatable," the patient can query their own history to find patterns, identify gaps in care, or prepare for a consultation in seconds.

The era of the vaulted file is ending, and in its place, we are seeing the rise of a decentralized, intelligent, and highly portable record that functions as the connective tissue for a person’s entire life. This is the final piece of the puzzle for the executive patient, and it is the catalyst that will force the rest of the healthcare ecosystem to finally prioritize the consumer over the institution.

The Path Forward: A Mandate for Leadership

Nothing is certain in the current landscape, and there are no easy ways to predict exactly how these trends will unfold. However, we are no longer looking at minor ripples. There are clear signals that the healthcare landscape is undergoing a fundamental restructuring, driven by the collision of AI technologies and the pattern shifts in human behavior taking place right in front of us.

It is now a strategic imperative for leaders across every part of the healthcare journey to understand how these changes will impact their business and their interactions with the consumer. The "Executive Patient" is already here, and they are looking for partners who can keep pace with their new expectations for data liquidity and transparency.

Change of this magnitude can be uncomfortable, but the future it promises is exciting. Organizations that continue to prioritize siloed experiences will find themselves increasingly invisible to the modern consumer. In order to be great in this new era, you have to figure out how to adapt. The winners will not be those who fought to maintain the status quo, but those who had the foresight to evolve alongside the people they serve.

In Part 1, you learned the big idea: ChatGPT is not one tool. It is a workspace with multiple modes, each built for a different job. This week is the create and understand layer. These are the tools that move you from quick chat answers to real-world usefulness: looking things up, doing deeper research when needed, making sense of products, speaking instead of typing, and working with images. Once you know when to use which mode, you stop fighting the product and start getting consistently better results.

Feature Index

Group 1: Find current information

ChatGPT Search

A mode that searches the web and answers with sources, so you can see where the information came from. Plain chat can be wrong or outdated. Search reduces guessing when you need current facts.

Where to find it: Open the tools list and choose Search, or ask a question that clearly needs current information.

When to use it:

• You want up-to-date info and sources.
• You want a quick answer, not a long report.
• You want to verify the sources yourself.

Group 2: Go deeper when the question is complex

Deep Research

Deep Research is a mode for complex, multi-step research. It searches the public web, can use files you upload, and produces a documented report with citations.

Is it available on Free?OpenAI says Deep Research is available on select plans in supported countries and territories, and usage varies by plan. If you do not see Deep Research in your tools menu, it is not enabled on your account right now.

Where to find it: Select Deep Research from the tools menu, then describe the task.

When to use it:

• The question needs many sources, not one.
• You need synthesis, not a quick fact.
• You want a longer answer with citations you can verify.

Search vs Deep Research

Use this simple rule:

• Use Search for a fast answer with sources.
• Use Deep Research for a thorough report with citations, when it is available on your plan.

Group 3: Make decisions with constraints

Shopping with Search

When you ask shopping questions, Search can return product-style results and comparisons.

When to use it

• You want a shortlist fast.
• You are still exploring options.

Shopping research

Shopping research is designed for deeper decisions where constraints and tradeoffs matter. It asks follow-up questions to narrow your needs.

Where to find it: Select Shopping Research from the tools menu, then describe what you want to shop for.

When to use it

• You have a real budget ceiling.
• You have must-haves and deal-breakers.
• You want the best for me, not the most popular.

Group 4: Talk instead of type

Voice conversations

Voice lets you have a spoken conversation with ChatGPT.  Voice lowers friction. It is useful when you want to think out loud and have an interactive session. Voice has usage limits and the experience can change over time. Voice can also make mistakes with names and numbers, so check important details.

Group 5: Work with images

Image understanding

You upload a screenshot or photo, then ask ChatGPT to explain it, summarize it, or help you interpret it. Free users have limited uploads per day, and limits can tighten during peak hours.

A simple personal-finance example: 

You screenshot a bank fee page because the wording is confusing.You ask: Explain this in plain English. Then list the fees I should watch for.

Image generation

You describe an image, and ChatGPT generates it. Images you create can be saved in your image library for reuse. Free tier image generation is limited. Simpler prompts usually work better.

My images library

A place where images you create are saved, so you can browse and reuse them without hunting through old chats.

One-screen recap

• Use Search when you need up-to-date info and sources.
• Use Deep Research when you need multi-step synthesis and a documented report
• Use Shopping with Search for quick product shortlists.
• Use Shopping research when constraints and tradeoffs matter.
• Use Voice to think out loud, but double-check important details.
• Use Images to understand screenshots and create visuals, with Free-tier limits.

🧾 Tax Deduction Finder

Most people either miss deductions they could claim, or waste time chasing ones they can’t. The trick is filtering: your job status + your filing situation determines what’s even possible. In 10 minutes, you’ll use ChatGPT + IRS sources to produce a clean “deduction shortlist,” the exact questions you need to answer to qualify, and a tracker to collect proof (so you’re not scrambling at tax time).

Step 1 — Identify your “tax profile” (2 minutes)

Write 5 bullets (rough is fine):

• Income type: W-2 employee / 1099 contractor / both / landlord / small business
• Filing status: single / married / HOH
• State: (helps for state-specific ideas, but we’ll focus on federal)
• Big life events this year: moved, new baby, bought/sold home, major medical, ducation, side hustle started, etc.
• How you track expenses today: none / bank app / spreadsheet / receipts

Important: If you’re a W-2 employee, most “unreimbursed employee expenses” are not deductible under current rules, with limited exceptions (certain specific worker categories). IRS

Step 2 — Have ChatGPT generate your deduction shortlist (3 minutes)

Copy/paste this prompt into ChatGPT:

Role: You are my “Tax Deduction Finder.”Goal: Create a realistic list of potential U.S. federal tax deductions/adjustments/credits I should check, based on my job/status.

My tax profile:

• Income type: ___
• Filing status: ___
• State: ___
• Life events: ___
• Expense tracking method: ___

Rules (must follow):

1. Do not guarantee eligibility. Treat everything as “possible—verify.”
2. Use IRS.gov as the primary source. If you mention a category, include the most relevant IRS publication/page name.
3. Separate into sections: Likely, Maybe, Unlikely/Not allowed (with reason).
4. For each item, provide:
   • What it is (1 sentence)
   • Who typically qualifies (1 sentence)
   • What proof to keep (receipts/logs/forms)
   • What would disqualify it (1 sentence)

Output format: a clean table.

Step 3 — Auto-create your “Deduction Tracker” table for Google Sheets (3 minutes)

Now paste this follow-up prompt:

Build me a Deduction Tracker I can paste into Google Sheets.Output CSV only (no commentary).

Columns:Category, Potential Deduction/Credit, Why It Might Apply (1 line), Eligibility Questions (3 short bullets), Proof to Keep, IRS Source to Verify, Est. $ Impact (Low/Med/High), Status (Researching/Confirmed/Not eligible), Notes

Fill 8–12 rows using the Likely and Maybe items from my shortlist. Leave “Est. $ Impact” as Low/Med/High (no dollar guesses).

Make the “Eligibility Questions” concise so I can answer them quickly.

Then:

1. Copy the CSV output
2. Open a blank Google Sheet → click A1 → paste
3. You now have a tracker that guides what to verify and what documents to save.

Step 4 — Do a 2-minute “IRS reality check” (2 minutes)

Pick your top 2 items from the tracker and verify them fast:

1. Click the IRS Source to Verify entry (or search the exact publication name on IRS.gov).
2. Confirm the key rule(s) and exceptions. For example:
   • Home office rules are very specific (see IRS guidance). IRS
   • Travel/car records require substantiation and good recordkeeping. IRS
   • Unreimbursed employee expenses are generally not deductible for most W-2 employees, with limited exceptions. IRS
3. Update your Sheet status: Confirmed or Not eligible.

This step prevents “TikTok deductions” from wasting your time.

The Payoff

You end with a personalized deduction shortlist that’s grounded in your job/status, plus a simple tracker that tells you exactly what to verify and what proof to keep. That’s how you avoid two costly outcomes: missing legit deductions and claiming shaky ones that create audit stress.

Transparency & Notes for Readers

• Free tools only: ChatGPT + IRS.gov (+ optional Google Sheets).
• Not tax advice: This workflow helps you organize questions and documentation; it does not replace a CPA/EA.
• Rules change: Always verify against IRS sources for the tax year you’re filing. IRS
• Educational workflow — not financial advice.