Interview with Ben Medlock
summary
Explains SwiftKey as a product and company that replaces thumb-typing frustration with an SwiftKey keyboard that uses an AI language model to predict words and correct touchscreen uncertainty; describes touch as introducing ambiguity that requires probabilistic models rather than rule-based autocorrect; outlines work toward gesture-based word creation and higher accuracy to reduce mistakes while preserving creative input, and frames broader AI goals as modeling human-brain flexibility rather than full consciousness.
AI-written summary · about
related talks
21mThe Impact of Word, Multiple Word, and Sentence Input on Virtual Keyboard Decoding Performance
10mDavid Boloker Wants to Replace Keyboards with Voice Commands
13mUniKey: Enabling Surface-Based Typing with Commodity Smartwatches via Cross-Modal Learning
25mAI-Assisted Swift in Gamifying an iOS App
03mInterview
34mWhen Software Understands the User
20mModelling Learning of New Keyboard Layouts
19mATK: Enabling Ten-Finger Freehand Typing in Air Based on 3D Hand Tracking Data