Voice Interfaces

Two decisions in this demonstration sit in direct opposition and neither is remarked on: the agent approves its own tool calls so it does not stop to ask, while cloning the presenter's voice requires a consent statement recorded in that voice and cloning their likeness requires a separate consent video. Maximum friction to copy a person, zero friction for the agent to act. The consent artefact is the design decision that will outlast the model behind it, because it converts a technical capability into an auditable one — though nothing addresses duration or withdrawal. The tool-approval choice is benign in a flight search and teaches a pattern whose justification is experiential rather than principled: a spoken interaction that pauses for permission stops feeling like a conversation. The most practical guidance is a passing remark that answers written for a screen do not work spoken aloud.

Microsoft Build

Three claims here are worth separating from the announcements around them. The speech numbers are meaningful in different ways: blind preference testing is the right evaluation for synthesis where no ground truth exists, and a latency under 150 milliseconds crosses the threshold where spoken exchange stops feeling like a request and starts feeling like conversation. The principles slide contains an actual position rather than decoration — capability should be learned rather than inherited, which takes a side in a live argument about building on existing base models versus training capability directly. And the training description carries the most transferable idea: generate solutions, score against verifiable ground truth, reinforce the better ones, with the clause about verifiability doing all the work. That explains the shape of progress across the field, and it is the same boundary Hassabis drew at Davos, arriving here as engineering description rather than forecast.

Microsoft Build

The clearest statement at I/O of how an agent API differs from a chat API, and the reasoning behind each departure is stated rather than assumed. Three changes matter. Conversation state moves to the server: a call returns an identifier, and passing it back continues the thread, retiring the client-side history array. The data model abandons alternating user and model turns for discrete steps, on the argument that a trace containing reasoning, tool calls, environment responses and compaction was never really a conversation and modelling it as one distorted it. And agents receive their own persistent remote environment rather than acting on the caller's machine — addressable by identifier, and shareable, so a research agent's output files become an application builder's input without passing through the context window. Schmid is explicit that scaffolded environment files are deliberately not model input, which is what keeps large artefacts out of the context budget. Schaeff's first half covers the real-time voice path, where the notable property is speech-to-speech across ninety languages with transcription of both directions.

Google I/O

The most revealing moment in this segment is not the hardware but a question the assistant asks before acting: would you like to confirm. On any other device that is unremarkable; on glasses it states the entire design problem. Every consumer transaction interface built in two decades has relied on a visual review step where errors get caught, and a device worn on the face has no review screen in any meaningful sense. What replaces it is the assistant restating what it is about to do, which must contain everything a user would have checked and be short enough to listen to — two requirements that pull against each other. Adding a tip after confirmation shows this is a conversation with amendments rather than a single transaction, which makes the boundary between negotiating and committing a design decision rather than an emergent property.

Google I/O

The sentence founders should sit with concerns cost structure: a technology this disruptive inverts the build-versus-buy calculation companies make. For thirty years that calculation was stable, and software companies existed in the gap between what a customer needed and what they could justify building. The concrete example is more useful than the abstraction — an interaction costing ten dollars means you could not afford the customer experience you wanted, which describes a category of product that was economically impossible rather than merely underserved. The observation with the widest implications is that operating in English addresses roughly ten per cent of the world, paired with the harder question of whether these systems handle a three-hour conversation. The analogy that does not hold is the internet, which created distribution where none existed rather than changing the cost of things already done.

World Economic Forum Annual Meeting