When Every Research Firm Uses AI: A Quiet Note on Reflexivity and Disruption

Exploring the Future of Industry Research in the Age of AI-driven Prediction

This piece follows an earlier reflection titled What’s Still Mine? A Knowledge Worker’s Quiet Question in the Age of AI.

There, I explored what it means to keep one’s voice in a world where machines can predict and produce so much.

In this note, I move from the personal to the systemic. I shift from that inner sense of disorientation to the broader implications for research itself.

It began with a simple question:

What happens when industry research and consulting firms widely adopt AI? Not just to organize data or identify trends, but to launch apps and build interactive platforms. What will this industry become?

This is not a forecast.

It is a quiet moment of sensing the future, when it presses closer and begins to shift the ground beneath us.

1.  What Products Will Future Research Firms Offer?

AI and platformization are reshaping the form of consulting services. Where we once delivered reports and slide decks, the offerings may soon look like this:

1.1  Insight-as-a-Service Platforms

Clients type in a question such as, “How will China’s restrictions on rare earths affect the EV supply chain?” The platform then generates data summaries, trend charts, cross-industry analysis, and strategic recommendations. These tools turn one-off reports into ongoing dialogues.

1.2  Auto-Generated Competitive Briefs

Clients input a competitor’s name and receive a ready-made briefing, including financials, market positioning, core strategies, and threat analysis. Output formats may include PDF, PPT, or direct integration into internal databases.

1.3  Semantic Monitoring Platforms

These tools track not just keywords but shifts in tone and intent. For instance, a system might detect how NVIDIA’s language around edge AI has evolved across earnings calls, and notify clients when new signals like “rising cost pressure” emerge.

1.4  Narrative-led Scenario Models

These combine AI with futures thinking. They help companies model multiple paths based on strategic narratives, such as: “If Apple stops developing its own AI chips, how will the supply chain reorganize?”

1.5  Analyst-as-a-Personality

Clients can choose which kind of analyst to interact with: a cool-headed strategist, a contrarian observer, or an East Asia industry expert. Each persona interprets data through a distinct frame of reference, offering a range of perspectives.

2.  How will This Market Evolve?

2.1  Short term (1–3 years)

Traditional report-based firms will face pricing pressure and delivery challenges. Companies with proprietary databases and engineering capacity will rapidly move toward platform and API offerings. Clients will increasingly favor real-time, interactive, and demand-driven insight platforms.

2.2  Medium term (3–5 years)

Analysts will evolve into prompt designers and content curators. They will:

• Help clients shape the right questions
• Design data extraction and response formats
• Translate technical output into human-centered strategic stories

Consulting value will shift toward strategic framing and cultural-context translation. Insight becomes a stylized product. Smaller firms without technical strength will rely on narrative and tone to differentiate.

2.3  Long term (5–10 years)

The traditional report delivery model will fade. Firms that fail to become platforms will be marginalized. Enterprises will build internal insight studios. External consultants will become embedded coaches. Independent analysts with a unique voice and framing may gain loyal followings.

3.  When Everyone Uses AI to Predict, What Happens?

This may be the most uncertain and most profound question.

When every firm, advisor, and strategist uses AI to predict others’ behavior, we enter the realm of reflexivity. This is not a technical flaw, but a logical paradox: once predictions become widely adopted, they start changing the reality they attempt to describe.

This idea traces back to George Soros’s theory of reflexivity. Market participants act on forecasts, and in doing so, reshape the market itself. The prediction becomes false by becoming true.

If everyone believes a stock will fall and sells it, it will fall because the crowd made it happen, not because the model was correct.

When AI models are trained on similar data and deployed to anticipate mass behavior, we may see:

• Strategy convergence and rapid saturation
• Trend bubbles inflated by self-reinforcing feedback
• Black swan events that no one is prepared for

AI has a blind spot. It can extrapolate from the past but:

• It does not realize it is altering the future it predicts
• It cannot grasp that publishing a forecast may change the behavior it observes
• It struggles with layered reflexivity: knowing that others know they are being predicted

4.  What Will This Do to Industry Research?

4.1  From Behavior to the Behavior of Predictors

Research will no longer center solely on “What will consumers do?” but instead ask, “When companies predict what consumers will do, how do they react and how does that reshape the market?”

4.2  Competitive Advantage Will Shift

The edge will not lie in who predicts best, but in who understands the bias and blind spots of dominant models.

4.3  Real Insight Will Come from Deviation and Renaming

“This market didn’t cool down. It overheated to the point that participants lost their agency.” That is not a line an AI is likely to generate. But a person can.

The role of the analyst will evolve from someone who observes trends to someone who observes how predictions are made, and eventually, someone who disrupts the model itself.

Conclusion: In an Age of Predictive Collapse, What Can We Still Do?

Individual behavior is unpredictable. Collective behavior once was. But when everyone uses AI to anticipate the collective, even that becomes distorted.

We are no longer studying markets. We are shaping them. The researcher becomes a participant, then a disturber.

The ones who remain won’t be those with the most accurate models, but those who can see when and why prediction breaks.

We won’t just write reports or output results. We may become designers of narrative, translators of context.

Insight will no longer mean knowing the most. It will mean knowing what still matters.

When everything becomes common sense, our job is to redefine what deserves our attention.

This note is not just about the future of industry research. It is about the quiet evolution of those who still care to ask: What is worth naming, when prediction becomes the norm?

This article is part of our Cultural Signals and Emerging Trends series.
It explores how subtle shifts in culture, behavior, and values, especially around work, identity, and technology, may quietly reshape the future.
These reflections aim to capture early signals, not as predictions, but as prompts for deeper understanding.

See more in this category, or explore more notes here.

Why AI Devices Are a Habit Revolution, Not Hardware

Generative AI is spreading quickly, yet few people actually use it on a daily basis. The issue is not about what AI can do, but about the habits that have yet to form. OpenAI’s collaboration with designer Jony Ive on a screenless AI device is not just about launching new hardware. It is an attempt to reshape how we live with AI, transforming it from a tool we turn on to something quietly present in our everyday routines.

The true shift is not in technical specifications or product form, but in how AI becomes a natural part of life, something we use without needing to think about it.

In this quiet but critical competition, the defining question is simple: who gets to shape the way we interact with AI? The answer may determine who leads the next generation of technology platforms.

When Jony Ive joined forces with OpenAI to create a screenless AI device, much of the attention naturally focused on what the product might look like. Would it resemble an iPod Shuffle? Could it redefine what an AI device is supposed to be? But the real question was never about its appearance. What truly matters is the problem it is trying to solve: how to move AI from a functional tool into something that becomes part of our everyday rhythm.

This is not just a hardware innovation. It is an attempt to reshape the relationship between humans and technology.

1.  Why Generative AI Feels Ubiquitous but Isn’t Yet a Daily Habit

As Benedict Evans observes , generative AI tools like ChatGPT have reached over 30 percent penetration in less than two years, a pace rarely seen in the history of technology. Yet one key data point reveals a striking disconnect. Only 5 to 15 percent of users engage with these tools daily.

In other words, we are witnessing the rise of a breakthrough technology that has not yet found a place in most people’s daily routines. It is something many know how to use but rarely think to use.

This is not a technical failure, nor a matter of missing features. It is a gap in habit.

2.  OpenAI’s Strategy: Making AI Feel Natural Instead of Just Smarter

OpenAI clearly recognizes this challenge. Its collaboration with Jony Ive is not about showcasing algorithms. It is about changing when and how people engage with AI.

The device they are developing is said to be screenless, lightweight, and wearable. It includes voice interaction and environmental awareness. Its ambition is not to make AI more impressive, but to make it less intrusive. There is no need to open an app, type a prompt, or consciously remind yourself to use AI.

Instead, AI begins to take the form of ambient technology, quietly present, always nearby, and available when you need it.

At the heart of this shift is a simple idea. AI no longer waits for you to find it. It lives alongside you.

3.  Why Habit, Not Features, Will Decide the Future of AI Platforms

History shows that with every new generation of platforms, the real contest is not about features. It is about retraining habits.

• We learned to scroll because the iPhone taught us.
• We learned to search because Google made it second nature.
• We learned to upload our lives because Instagram turned it into muscle memory.

Now, OpenAI is trying to do something similar. It wants asking AI to become an unconscious reflex, something people do without thinking about it.

This is the real strategic goal. It is not about outperforming other AI models. It is about becoming the default entry point for everyday use.

4.  Designing Post-Screen AI Interfaces Around Meaning and Presence

The screenless design of this device is not a rejection of technology. It reflects a belief in what comes after the screen.

AI no longer needs a display to show you what it knows. It can listen, understand, and respond. It can work quietly in the background, even when you are not paying attention.

This introduces a different logic for how we interact:

• From swiping and tapping to speaking and sensing
• From visual cues to semantic understanding
• From choosing from menus to having your intent anticipated

The screen era was built around choice. The AI era is shaping itself around understanding and presence.

Conclusion: The Shift Toward Ambient AI and Unconscious Use

OpenAI’s device is not a hardware revolution. It is a first move in the design of everyday habits. The goal is not to impress you. The goal is for you to forget you are even using it.

This is not a race to build the fastest algorithm or the most advanced chip. It is a race to define the behavior people repeat every day with AI.

OpenAI is not challenging the market with specs. It is shaping the future through habit. The real question is simple. Do you reach for your phone, or do you speak to your AI? Do you scroll Instagram, or do you ask for what you need?

When AI begins to live within the details of our lives, from our sounds and gestures to our routines and even our pauses, the competition among platforms will no longer be about apps and devices.It will be about who quietly takes root in our unconscious habits.

This revolution is not in our hands. It is in our every day.

This article is part of our Cultural Signals and Emerging Trends series.
It explores how subtle shifts in culture, behavior, and values, especially around work, identity, and technology, may quietly reshape the future.
These reflections aim to capture early signals, not as predictions, but as prompts for deeper understanding.

See more in this category, or explore more notes here.

What’s Still Mine? A Knowledge Worker’s Quiet Question in the Age of AI

In an era where even industry research may be reshaped by AI, I found myself asking: If my way of thinking and working can be replicated, what’s still mine?

This is a quiet reflection from someone doing knowledge work, about self-doubt, about trying to find a personal rhythm again, and about what it means to have a relationship with thought.

I don’t dislike AI.

In fact, I kind of like it. It helps me organize research ideas, and finishes in minutes what would have taken me hours, maybe even days, on my own.

But lately, I’ve started thinking about something: If it can learn my research methods, understand my writing rhythm, and even imitate the way I process ideas, then what’s left that’s still mine?

Especially when even “research,” the one thing I once believed to be the most personal expression of my thinking, becomes something that can be predicted, modeled, and reproduced.

I know it isn’t my enemy.

But more and more often, after finishing an insight piece, I catch myself wondering quietly: “Was that me? Or could it have written this just as well?”

I’m not trying to make a point. I just want to ask.

And maybe I’m simply trying to find my own pace again. A way of expressing that still leaves a trace.

Sometimes, it does the work so well that it leaves me stunned.

I’ve spent days editing the logic of a single piece, and it can generate multiple clean drafts in seconds, well-structured, precise, and sometimes even clearer than mine.

And in those moments, I start to think: If I feed it enough material, enough context, enough samples of my tone, could it become “me,” more efficient, more stable, more consistent than the real thing?

It’s not a sad thought. It’s more like a quiet sense of fading.

It’s like the methods I’ve spent years developing aren’t irreplaceable after all. They’re just another process that can be optimized.

And I, perhaps, have become a replaceable node.

But eventually, I realized something: AI can replicate results, not the path that led to them.

It can copy research logic, predict how I might structure an argument, generate paragraphs that look like mine.

But it has never sat through my long stretches of uncertainty, those hours of rewriting, the quiet questioning of whether a certain point feels too soft or too harsh, too early or too late.

It doesn’t read a sentence and suddenly recall a three-year-old technical question.

It doesn’t arrive at a closing thought and feel the urge to go back and restructure the entire piece from scratch.

It doesn’t feel doubt.

It doesn’t pause.

It just moves forward.

And maybe that’s exactly what I still need to protect, not just the work itself, but my relationship with thinking.

That strange, meandering process of writing while not yet sure, questioning while typing, slowly clarifying as I go.

Maybe that’s what being a researcher really means.

So instead of asking what’s still mine, maybe I’m trying to find a reason to stay, a reason to keep understanding this world in my own way.

I know that research work, especially forecasting or strategic analysis, will feel more and more like an uneven race against the models.

But I also know this: Human value lies not in speed, but in the willingness to sit with uncertainty.

AI will keep moving forward.

And I might walk slowly.

I might take detours.

I might break down.

But as long as this path still has space for the kind of observation that’s made of hesitation, emotion, and doubt, then maybe, just maybe, there’s still a place for me.

AI might be able to analyze shifts in supply chains. But it doesn’t understand what a manager’s silence means when a plant is shutting down.

Right now, I still have the ability to see the layer of uncertainty AI can’t.

So I want to say, even if I’m not in front, that’s okay.

I still choose to leave a trace.

Even if it’s a small one.

As long as it’s one I walked myself.

This isn’t a piece with clear answers, nor is it a prediction about the future.

It’s simply an honest question: What, if anything, is still mine?

I wrote it to remember the unease and the thinking that existed in this moment, a quiet trace of a time when I wasn’t sure, but still wanted to understand.

This article is part of our Cultural Signals and Emerging Trends series.
It explores how subtle shifts in culture, behavior, and values, especially around work, identity, and technology, may quietly reshape the future.
These reflections aim to capture early signals, not as predictions, but as prompts for deeper understanding.

See more in this category, or explore more notes here.