Why Top Innovation Ideas Stay Buried And How AI Can Help Surface Them

Forbes

19 hours ago

Ludwig Melik, President of Operations North America and CRO at HYPE Innovation, the Smart Innovation Platform.
I once grabbed a handful of mixed nuts during a brainstorming session when someone joked, 'Why are the Brazil nuts always on top?' That throwaway comment sparked a conversation, not about snacks, but about how ideas rise (or don't) inside organizations.
It turns out, there's actual physics behind it: The 'Brazil nut effect' explains why, in a shaken container, larger objects migrate upward while smaller ones settle below. The metaphor applies too well to innovation: Some ideas, not always the best, rise to the top, pushed by volume, confidence or simple familiarity.
And the best ideas? Too often, they stay buried beneath the noise.
Loud ideas win by default, but they're often the wrong ones.
If you've worked in product or innovation teams, you've seen it happen. The quiet insights get overshadowed. The unusual angles go unrecognized.
Unless your system is intentionally designed to surface and evaluate those quieter ideas using smart filtering, contextual comparisons and asynchronous input capture, the signal gets drowned out by the noise.
That's where AI can help. It doesn't replace human judgment. It reshapes how the shake works and what rises as a result.
Innovation isn't linear, so stop managing it like a project plan.
Veterans in innovation know the process has evolved drastically over the years:
• 1990s: Innovation was confined to R&D. Slow, expensive and often disconnected from real-world feedback.
• 2000s: The rise of 'innovation labs' and crowdsourcing tried to democratize innovation. Some success, much theater.
• 2020s: Agility went from buzzword to baseline. Real-time feedback, iterative prototyping, multilingual collaboration and cross-functional visibility became essential.
Now we're entering a new phase: AI as co-creator. It's not a visionary, but a force multiplier and an always-on filter for what matters.
AI can make innovation work better.
AI's most valuable contributions to innovation don't come from grand predictions or flashy prototypes. They show up in quieter, more practical places—the kind seasoned innovation professionals will recognize immediately.
Here's where the real value lies:
AI can assist contributors as they formulate ideas, offering real-time suggestions to clarify, strengthen and structure submissions. This early guidance reduces noise and raises the baseline quality of contributions. It's not automation replacing creativity. It's scaffolding that helps good thinking rise.
In mature systems, AI supports evaluation by summarizing submissions, attachments and discussions. This isn't about automating judgment. It's about accelerating clarity.
When AI clusters similar or duplicate ideas, it leads to sharper synthesis and fewer blind spots.
Global innovation often breaks down over language or inconsistency. AI-driven translation and summarization reduce that friction. True inclusion isn't about access. It's about comprehension. AI helps ensure great ideas don't get lost in translation.
AI handles the background tasks like tagging, structuring, formatting and routing content. This frees teams to focus on insight, not administration. Let the machine do the grunt work. Let your people do the thinking.
Your innovation pipeline is broken. Here's how to fix it.
Too many organizations treat innovation like an assembly line: Feed ideas in and expect validated solutions out. But that model ignores the messy, emergent nature of creative work.
A more useful metaphor: a garden.
Ideas sprout and compete for sunlight. Some die off. Others cross-pollinate. Serendipity and timing matter as much as intent.
AI acts like a sensor network and irrigation system. It helps you notice overlooked growth and keep things moving. But it's still on you to prune, guide and cultivate.
Practical fixes can surface high-impact innovation ideas.
To bring better ideas to the surface—not just the loud ones—here's what experienced teams can do:
Great ideas often come after the meeting ends. Use structured digital spaces to capture inputs before and after live sessions, especially from quieter team members or those who may prefer contributing in writing in their own language.
Good thinking doesn't always show up on schedule, so stop scheduling all of it.
Shift from popularity contests to criteria-based evaluations. Combine AI-assisted filtering with human scoring to assess impact, feasibility and alignment. The most-liked idea isn't always the most valuable.
Let AI surface clusters, duplicate ideas and emerging trends. Then bring in cross-functional teams to interpret them. AI sees the patterns. You assign the meaning. That's the real collaboration.
AI won't replace you, but it might stop you from ignoring the right idea.
Innovation will always be human. AI offers leverage, not leadership.
Think of it as the most tireless intern you've ever had: fast, context-aware, multilingual and never on vacation. But that intern doesn't know your strategy, your customer or your culture. You do.
So maybe the question isn't, 'How do we stop the wrong ideas from rising?' Maybe it's, 'How do we make sure the best ones aren't ignored?'
Because when the system is right, when tools and teams align, great ideas don't rise by luck. They rise by design.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?