Beyond the Buzz

Every new wave of technology (and this is a big one!) brings a tidal rush of headlines, predictions, and hype - none more so than AI. This week, we begin our series “How We Design AI and Not Vice Versa” with a reality check: what AI can deliver right now and what’s still beyond today’s capabilities.

AI has become a cultural juggernaut - fuel for viral headlines on social media, dinner table debates, investment frenzies (“don’t miss out!”), and existential panic about our jobs and the safety of our children. We see sci-fi robots stealing the spotlight in movies, read about algorithms allegedly plotting to take over entire industries, and hear that AI will either destroy or save the world. Yet somewhere between the hype and reality lies the real question: what can AI actually do for us today, and where’s the fine line between genuine innovation and smoke-and-mirrors illusion?

Experience map, showing the summary of user interviews in several grid layers, including actors involved, experience feedback, and others.

AI: Separating Science Fiction from Everyday Use

Narrow AI dominates: Today’s most useful AI systems are “narrow” AI - designed to perform one task extremely well, such as spotting fraudulent transactions or translating languages. They don’t understand context, and they cannot transfer skills. (IBM and Nature)
Sci-fi dreams vs. real constraints: General AI - the kind that could reason, feel, or make decisions across any domain without human guidance - does not exist yet. Most experts believe it’s decades away. (Washington Post)

If the average LinkedIn feed is to be believed, AI has already mastered everything from human emotion to poetry, creativity, and a lot of things we humans don’t want to give away. Yet, away from the buzz, most AI systems in use today are narrow and distinctly non-magical (sorry enthusiasts). The real breakthroughs are far more subtle - think predictive text, smart email filters, instant product recommendations, and voice assistants. These are powerful - but their intelligence is specialized, not general.

Useful AI Is Everywhere - But It’s Not Magic

Artificial Intelligence is seamlessly integrating in our daily lives, enhancing user experiences, streamlining business operations, and even aiding in creative workflows. While these advancements are significant, they often manifest in subtle, incremental improvements rather than groundbreaking revolutions. For example:

AI Voice Assistants

What It Is: AI powers digital assistants like Siri and Alexa, helping with reminders, smart devices, and queries.
Why It’s Useful: Offers hands-free convenience and quick access to information.
Limitations: Still relies on scripts and struggles with nuanced conversations.
Future Outlook: Ongoing advancements aim to improve contextual understanding and enable more natural, human-like interactions.

Sales Agents

What It Is: HubSpot’s AI tools, like Breeze Assistant, help with content creation, meetings, and customer management using CRM data.
Why It’s Useful: Automates routine tasks, prioritizes leads, and frees teams for strategic work.
Limitations: Complex decisions still need human oversight; effectiveness depends on data quality.
Future Outlook: More sophisticated automation and deeper integration will enhance personalization and efficiency.

“The best AI experiences are not (and will not) be built by machines alone - they are co-created with people."

Where AI Falls Short (And Why That Matters)

The flip side of every innovation story is what AI cannot do - or what it does poorly when expectations exceed today’s capabilities. Understanding these limits is critical for ethical design, transparency, and human-centered AI development.

Context and common sense: AI lacks genuine understanding of context. A chatbot might ace a product FAQ but fail utterly at a complex, nuanced negotiation. Creative tools generate content, but their “imagination” is only as deep as the data they’ve been fed.
Critical thinking and judgment: AI excels at recognizing patterns but struggles with critical thinking, ethical judgment, and social empathy. For example, research shows that over-reliance on AI tools can erode critical thinking skills, leading to narrower perspectives and biased analyses because AI lacks the capacity to evaluate assumptions or reason deeply.
Bias and reliability: AI learns from data - and data reflects human worldviews, histories, and errors. If the training material is skewed or flawed, the system will be too. This isn’t just a bug or a “hallucination”; it’s a core challenge of AI development that demands attention at every stage.

AI’s impact varies across industries, and despite high hopes, many technologies are still in progress. Healthcare chatbots can support diagnosis but often require human oversight due to risks of inaccuracies and lack of empathy, as studies highlight concerns over over-reliance on AI-driven advice (JMIR, Standford). Similarly, generative AI like ChatGPT and Gemini can produce content but are prone to errors and misinformation, serving as tools to augment human creativity rather than replace it entirely.

The Human Role: Designing AI for Real Value and Looking Forward Beyond the Buzz

The best AI experiences are not (and will not) be built by machines alone - they are co-created with people. We believe AI must extend human capability, amplify human judgment, and serve human goals. This isn’t just good design; it’s an ethical necessity. Product teams need to move beyond “Can we build it?” to ask “Should we?” and “For whom?” Every AI initiative must be grounded in real human needs, ensuring it solves genuine pain points rather than chasing novelty. Systems must be transparent, accessible, and aligned with user values - because without understanding how decisions are made, people can neither trust nor correct AI. And above all, AI should empower people by elevating their agency, not diminish it.

Human-centered AI (HCAI) bridges technology and meaningful impact by embedding ethics, empathy, and inclusivity at every stage of design. Unlike traditional AI that focuses on automation and efficiency, HCAI prioritizes human needs to create explainable, fair, and adaptable systems. While these changes are incremental, the trajectory is clear: the future belongs to AI that adapts to humans - not the other way around. The most successful products in the coming years won’t just be smart; they will be thoughtful, transparent, and deeply human. Design becomes our most powerful tool for shaping this future.  

AI’s story is still unfolding. Tomorrow’s breakthroughs could surprise us as much as today’s limitations. The real game-changer is designing AI through a human-centered lens, separating hope from hype to build responsible, delightful, and impactful solutions. The future shines bright if AI is built to elevate human potential, not replace it. True partnership between humans and AI requires integration, transparency, and mutual respect. Our unique creativity and collaborative spirit will thrive alongside AI, transforming the next leap into an ethical, imaginative, and profoundly human journey.

In our series “How We Design AI and Not Vice Versa: Conversations on AI, Ethics, and Designing Tomorrow”, we’ll journey beyond the hype to explore what AI can do today, uncover the ethics and biases behind algorithms, and reveal how human-centered design, UX, and responsible innovation come together each week to shape technology that truly serves real people.

Want to explore how AI, UX, and ethical design shape tech for real people? Visit www.yellowumbrella.design for our insights, practical resources, and inspiration for designing a better tomorrow.