For years, deep work meant silence. Closed door. Headphones on. Minimal noise. The ideal engineer was someone who could sit quietly for hours and produce.
But AI is starting to change what that looks like. As voice interfaces improve and models become more conversational, the act of building software and prompting LLMs is changing the dynamic of how we work. Developers are explaining their reasoning out loud, walking through logic verbally, and letting the models respond in real time. How will this impact office life is a question we’re curious to ask.
Why Voice Prompts Are Becoming More Widely Used
Typing introduces friction. Before something appears on screen, it has to be filtered. You decide how to phrase it. You structure it. You remove the rough edges. That editing step slows down raw thinking.
Speaking back and forth with an LLM via voice prompts changes this dramatically. When someone talks through a bug or an architecture idea, they don’t wait for perfect phrasing. They think in motion. They revise mid-sentence. They clarify as they go.
AI handles that surprisingly well. Instead of carefully drafting a prompt, developers can explain context naturally and let the model refine alongside them. The interaction becomes iterative rather than transactional. That changes behaviour.
The Office Problem
Most modern offices were built around the assumption that focus is quiet. People typing. Occasional meetings. Controlled background noise. But conversational workflows with AI tools don’t fit neatly into that environment.
Talking through logic for extended periods feels disruptive in a shared space. Even if no one complains, there’s social friction. People instinctively lower their voice, shorten their explanations, or revert to typing.
That hesitation matters. It interrupts flow. If AI increasingly rewards audible reasoning, then silence may no longer be the optimal default.
Why Remote Might Have An Advantage Here
Remote work changes the equation here. A private environment allows people to externalise their thinking without constraint. They can narrate freely, restart mid-thought, argue with the model, and iterate without worrying about disturbing anyone.
This may be an underappreciated productivity factor. The advantage of being able to work from home is not only flexibility or commute savings, it’s freedom to be able to work in a place and environment that works best for you, but increasingly also for your colleagues if we focus on the ability to treat AI as a conversational collaborator rather than a text-only tool.
If that mode of working becomes standard, your working environment has an even bigger impact on your productivity and equally the productivity of your colleagues.
w
Bottom Line
We may be redefining what focus means. In the past, focus meant shutting the world out. Increasingly, it may mean engaging deeply with intelligent tools.Â
As AI becomes more capable in real-time dialogue, the highest-performing engineers might not be the quietest ones. They might be the ones most comfortable externalising their reasoning and vocalising it out loud.
The leadership question shifts as well. Instead of asking how to restore traditional office norms, companies may need to ask whether their environments support the way AI-native work actually happens. Deep work is no longer just about silence. It may be about conversation.
How do you think this will impact the design of offices and the flexibility given to AI talent to work remotely? Let’s see, but this is a perfect example of 2nd order consequences from new technologies.
Donal Brady is Co-Founder and Head of Product at Work From Anywhere, a platform to help companies execute a successful workation policy. He has deep expertise and experience in the world of remote work having spent nearly twenty years working in finance leadership roles with global multinationals like PwC, before entering the world of global mobility with International SOS and then pivoting into the software industry with Accel-KKR backed Smart Communications. He walks the walk with Work From Anywhere, speaking 5 languages, having travelled to 65+ countries and worked in 10.
