๐ Thrilled to share our paper has been accepted @acm_chi !
"Can AR-Embedded Visualizations Foster Appropriate Reliance on AI in Spatial Decision Making?"
We conducted an empirical study (N=32) comparing AR see-through vs. 2D minimap in time-critical, AI-assisted spatial tasks.
Key finding: Contrary to expectations, AR embedded visualizations led to greater over-reliance on AI โ driven by perceptual challenges, proximity illusions, and heightened trust in "embodied" AI cues.
Yet AR still showed strengths in spatial mapping & egocentric reasoning.
Proud to contribute as second author alongside and am deeply grateful for the opportunity to collaborate with such a brilliant team.๐ @CartonLiu
TongyuNie, @EvanSuma , Victoria Interrante, and @ZhutianChen
Please check our paper here:
๐ท arXiv: https://t.co/yvr4ngBdS2
#CHI2026 #CHI
you know what
all of these "which is better" polls are silly
use codex or claude code, whatever works best for you
i am grateful we live in a time with such amazing tools, and grateful there is a choice
For AI-native:
In the future, setting menu should only be the fallback solution for an APP.
All settings and functions should be able to work/be modified in a CLI or Chat interface, directly or with a small artifact generated by AI.
Any (to c) apps letting the users jam in the setting menu should be discarded.
Introducing Lovable General Tasks ๐ฅ
You can now build way more than just apps:
โ Analyze data and generate PDF reports
โ Create marketing assets for your product
โ Build interactive pitch decks
โ Produce animated launch videos
I am personally so excited about this. Lovable is becoming the entire startup stack in one interface.
Build your product, design your pitch deck, create your launch video. All in the same place.
The team has been relentless. We're just getting started.