Hi there, This is the thirty fifth edition of D&T Special, a more in-depth view of topics that interest the Canvs team. Today's topic - How AI Can Help With Testing Products Before Accessing Real Users.
Note: If you are new here, we'd strongly recommend subscribing to the newsletter so you don't miss out on future updates. You can subscribe here.
✍️ From the Canvs Research & Editorial Desk
AI is now a critical tool that the product developers have at their disposal for testing products before they even reach real users. Agentic AI, a newer development in this space, goes beyond routine automation by performing tasks that once required human oversight. Whether it's reviewing prototypes or identifying potential issues in user flows, this technology helps designers catch problems early.
This week the Canvs R&E team has spent some time pondering this concept, let’s dive into some details.
How AI can help testing products before accessing real users
By shifting from human-led testing to AI-assisted testing, the design process becomes more efficient. These AI tools can simulate user journeys and flag issues like broken interactions or confusing navigation paths before a single human tester gets involved. It’s like having a built-in safety net, allowing designers to refine products sooner and more effectively.
### Key takeaways from this read:
#### 1. AI provides early validation
Tools like Maze allow designers to simulate user journeys and spot potential friction points in the flow before any real users interact with the product. It flags issues like unclear navigation or broken interactions which gives designers an objective assessment of how well their product holds up.
#### 2. It excels in handling technical checks but leaves room for human creativity
While agentic AI is excellent at spotting functional issues—like layout misalignments or poor button placement—it struggles with subjective aspects such as visual aesthetics and emotional design. The tool can recommend changes for usability, but its suggestions are based on patterns, not the nuanced insights a designer brings.
#### 3. It helps reduce costs by automating early testing
AI’s ability to handle early-stage testing reduces the need for human testers in the initial phases, which can often be time-consuming and costly. Teams save money on extensive manual testing. This allows them to reserve budget and resources for refining the product later on.
📰 In other news
Hackers launder most of Bybit’s stolen crypto worth $1.4B
AI transparency in UX: Designing clear AI interactions
Melbourne start-up launches 'biological computer' made of human brain cells
Figure's Humanoid robots now sort, pick and handle packages at human speed
OpenAI has released a research preview for their latest model GPT‑4.5 this past week
✨ Product find of the week
FLORA - Every creative AI tool, thoughtfully connected
This tool will help you create at the speed of thought and let you ideate, iterate, and explore faster than ever before. All of the top text, image, and video AI models in one infinite canvas.
Some highlights from the past month of D&T
Meta plans to link US and India with world’s longest undersea cable project
How Netflix built a distributed counter for billions of user interactions
And that's the lot! Thanks for checking out what we had to share with you this week, we shall catch up with you next Wednesday. Incase you aren't subscribed to the newsletter, you could subscribe here.
Also, please do let your friends know about D&T if they'd be interested! Canvs Club
