Most corporate foresight programs collect signals nobody acts on. Reports get filed. Trend radars get presented. Executives nod. Then, strategy teams keep building the same five-year plan they would have built without any of it. Intel runs corporate foresight differently, and John Miranda explains how.
The result is a foresight function that costs real money and changes nothing.
Intel runs corporate foresight differently. John Miranda, who has led Intel's Corporate Foresight Program from inside the Data Center and AI Strategy Office, built the program around one principle: insights without decisions are noise.
This article breaks down how Intel does it, what most companies get wrong, and how generative AI is rewriting scenario planning in 2026. It is written for strategic foresight managers, heads of innovation, and VPs of innovation who need to turn weak signals into resourced decisions across an entire organization.
John joined us on the Innovation Rockstars Podcast to critically examine how to enable corporate foresight in a global company.
🎬 The full video podcast or 🎧 pure audio enjoyment on Spotify or Apple Podcast.
Why corporate foresight breaks down inside global companies
Three patterns kill corporate foresight programs before they produce value.
John Miranda has named the underlying cause directly: "Without a doubt, losing trust" is the fastest way to kill a foresight program. Trust collapses when foresight teams overpromise certainty, miss commitments, or report findings that never translate into action.
The fix is not better tools. The fix is treating corporate foresight as a system of relationships, not a system of reports.
What corporate foresight actually does
Corporate foresight is the organized practice of identifying early signals of change, building plausible futures around them, and using those futures to inform strategy.
It is not a prediction but preparation.
A working corporate foresight function delivers four things to the business: an early warning system for emerging trends, a structured way to explore possible futures, a scenario planning capability that links to strategy, and a continuous learning loop that updates as new signals appear.
The discipline draws on futures studies and decades of scenario work at Shell, Intel, and others. But it only earns its keep when it changes a decision the business would have made anyway. Detecting an emerging trend is the easy part. Forcing a contingency plan into the strategic plan is the hard part.
Generative AI has expanded what is technically possible. The bottleneck is no longer signal detection. It is signal interpretation and the courage to act on weak signals before they are obvious.
Inside Intel’s corporate foresight program
Intel's program is built around three elements that work together. Each one targets a failure pattern that most foresight programs fall into.
A federated contributor network
Intel does not centralize corporate foresight in a single team. The program is a community of contributors drawn from across the business: engineers, strategists, market analysts, and sustainability leads. Miranda's job has been to identify employees who show intellectual curiosity and thought leadership, map their expertise, and create a go-to network for different subject matters. This solves the isolation problem at the root.
Miranda came to this work as a self-described "computer geek." A computer science degree from Michigan and an MBA from Texas led him into Intel, then into leading a team of application developers. Thus, the program is run by someone who knows both the technology stack and the business, not by a dedicated futurist disconnected from either.
A validation cadence
Members meet a few times each quarter to interpret findings, debate implications, and construct arguments for why a signal matters. Intel uses tools to support this, including the ITONICS Innovation Management Software for structured signal evaluation. The point of these sessions is not consensus. It is to pressure-test hypotheses before they reach senior management.
Executive partnerships
The third element is where most programs fail. Miranda has explicitly partnered with groups like the Corporate Strategic Office because, as he puts it, "coming up with insights without the accompanying change is not enough." Insights flow into real strategy conversations. Resource reallocation follows. Without this link, even the best scenario planning produces nothing.
A useful detail from Miranda: Intel's founders Gordon Moore and Andy Grove modeled the cultural posture the program now relies on. Both flew economy. Both drove subcompacts on campus. That modesty translated into a culture where intellectual contributions from any level get taken seriously, which is exactly what a federated corporate foresight program needs.
The Intel program treats foresight as a never-ending learning process. New tools, new methods, and new contributor relationships get added as the business context changes.
Four drivers of change that Intel's foresight program is tracking
A corporate foresight program is only as good as the signals it surfaces. On the Innovation Rockstars podcast, Miranda named four driving forces shaping Intel's strategic thinking. They are useful as a working example of what a foresight program should be producing for any large business.
Notice the pattern. None of these are predictions. Each is a key driver with multiple plausible outcomes, each one demanding contingency plans rather than a single bet.
Scenario planning that holds up under certainty
Scenario planning fails when it produces a single "most likely" view. It works when it produces a small set of plausible futures and a contingency plan for each.
A practical scenario set covers four future types:
Building three to four plausible futures, plus one preferred future, gives a strategy team enough range to test decisions without drowning in complexity.
But in the end, each scenario needs three things to be useful:
-
an explicit set of driving forces,
-
a clear narrative of how the world gets there, and
-
a contingency plan describing what the business would do.
Without the contingency plan, the scenario is a thought experiment. With it, scenario planning becomes a decision-support tool.
How generative AI sharpens deeper insights from weak signals
Generative AI has changed the economics of corporate foresight in two specific ways.
Signal detection scales. What used to require a team of scouts can now be done continuously across thousands of sources (Exhibit 1). Large language models can read patents, scientific papers, startup funding announcements, and policy documents at a volume no human team could match. Emerging technologies and weak signals surface earlier.
/Still%20images/Element%20Mockups%202025/foresight-stay-in-the-loop-2025.webp?width=2160&height=1350&name=foresight-stay-in-the-loop-2025.webp)
Exhibit 1: Prism alerts on interest increase for emerging trends
Signal interpretation improves. Generative AI can cluster related signals, surface indirect effects, and propose connections between unrelated developments. This produces deeper insights than manual analysis, especially when the signal sits at the intersection of two fields like synthetic biology and computing.
The risk is the same risk as every other AI application: confident output that is wrong. Generative AI is excellent at producing plausible-sounding analysis. It is bad at flagging when it is uncertain. A corporate foresight team that takes AI output at face value will fabricate trends.
The working pattern is human-in-the-loop. AI handles the volume problem. Humans handle the judgment problem. Big data narrows the field. Expert contributors decide what to act on.
ITONICS customers use Prism to apply this pattern at scale (Exhibit 2). Signals get detected and clustered automatically. Domain experts validate, score, and tie them to portfolio decisions. The result is a continuous learning loop instead of a quarterly project.
Exhibit 2: Prism identifies opportunities matching strategic criteria by continuously scanning 50M+ patents, publications, and market signals
Five rules to enable corporate foresight in your organization
A practical framework, drawn from how Intel and other mature programs operate (Exhibit 3).
Exhibit 3: Five rules to enable corporate foresight in your organization
These five rules will not produce immediate results. A corporate foresight program needs 12 to 18 months to earn credibility with executives and another year to influence resource allocation. The investment is small. The compounding return is large.
How ITONICS powers corporate foresight at scale
ITONICS gives corporate foresight programs the infrastructure to operate as a system instead of a series of projects.
The platform handles the signal-to-decision flow that most programs run on spreadsheets and slide decks. Weak signals get captured, clustered, and scored. Emerging trends get linked to the business context they affect. Scenario planning workflows let teams build and compare plausible futures. Portfolio views connect foresight insights directly to active initiatives and investments.
Prism handles the volume problem. It scans signals continuously, surfaces patterns, and identifies opportunities the team would otherwise miss. Domain experts stay in the loop for judgment calls. The entire organization gets visibility into the same signals, scenarios, and decisions.
For strategic foresight managers, this collapses two persistent problems: the disconnect between foresight and strategy, and the credibility gap with senior management (Exhibit 4). Both get solved by making foresight legible to the rest of the business.
Exhibit 4: Transform how your organization discovers, evaluates, and acts on emerging technologies
Intel, Toyota, DB Schenker, and other ITONICS customers use the platform to run corporate foresight as governed infrastructure. Not as a side project.
