AI’s Hype Cycle Meets Product Reality
There is substance behind the AI buzz, but for many teams the harder question is figuring out where that substance actually lives. With new models, features, and research dropping constantly, staying focused on difficult product questions is a challenge. That’s precisely why this moment favors teams willing to experiment and iterate through the noise.
This analysis draws on a Figma survey of more than 1,800 users—designers, executives, and developers—conducted between February 26th and March 3rd, 2024 across seven countries (US, CA, AUS, UK, JP, FR, DE), plus insights from two livestream discussions on the findings.
Developers Are Further Along Than Designers
When it comes to hands-on AI adoption, developers are ahead. They are 60% more likely than designers to report that AI has already transformed the products they work on. That gap isn’t necessarily about enthusiasm; it’s about fit. Developers use LLMs for daily tasks like generating starting points and translating between languages, tasks where AI output is proving reliably useful.
Designers, by contrast, spend much of their foundational work on nonlinear exploration—building deep understanding of user needs and problem spaces—which current AI tools don’t influence as directly. AI can turn a mockup into code, but it doesn’t yet carry much weight in the fuzzier front end of the design process.
Still, interest in AI isn't confined to design teams. One livestream attendee noted that their last three AI projects originated from stakeholders outside design, such as programmers and subject matter experts.
Individual Gains, Unchanged Collaboration
The survey shows AI has shifted individual workflows far more than collaborative ones. Eighty-five percent of respondents say AI has impacted their personal productivity, typically through text and image generation or by acting as a sounding board. Respondents were three times as likely to report significant transformations in their individual work versus their team's work. Groups haven't yet seen AI change coordination tasks like alignment or meeting facilitation.
One webinar participant, James, challenged that framing: if an individual is "collaborating" with an AI, is that work truly individual? The distinction matters. If AI product builders focus solely on personal productivity gains, they risk missing the larger opportunity. Products that are genuinely transformational, the report suggests, will need to address how teams work together, not just how individuals work alone.
Transformational AI Needs More Than Tech
Expectations for AI-driven change vary sharply by sector. Respondents in technology (41%), professional and business services (40%), and retail (39%) anticipate major shifts in their products within the next year. But those in healthcare, energy and utilities, and telecommunications expect much less near-term impact—even though those industries underpin daily life. Realizing AI's full potential means solving problems in these slower-moving fields as well as in software.
The Pressure to Ship Versus the Path to Success
Leaders eager to capitalize on AI are pushing teams to ship quickly, often at the expense of thinking and experimentation time. One attendee described the result as "a race to the bottom," with everyone rolling out AI features in the same way. Another cited AI feature fatigue within their own team, noting the flood of available tools with few established best practices.
The survey data suggests a stark split in outcomes. Only one-third of all respondents are proud of the features they’ve launched so far. But among those building AI into the core of their product, that figure jumps to 82%. This group reports that generative AI has made their products more valuable and useful—but getting there hasn't been easy.
Zan Gilani, a Principal Product Manager at Duolingo, offered a concrete case. His team built eight generative AI products last year that didn't land before the ninth one—conversational speaking with an AI partner, now in beta—found its footing. The differentiators, he says, were rapid iteration, lightweight prototyping, and a clear tie to a real user need.
The Real Winners Will Be Iterative
Not every hyped technology proves transformative, at least not in the way predicted. Those that do share a common thread: teams that see long-term potential and commit to experimenting until they get it right. That kind of work demands tolerating failure and pushing back against disillusionment while shipping anyway. The teams who do that, the report concludes, may find the outcome was worth the effort.
