This content originally appeared on Opera News and was authored by Krystian Kolondra
By Krystian Kolondra, Opera
Last week, I watched my teenage son spend 47 minutes booking a simple flight. He toggled between 12 browser tabs, compared prices across six sites, checked baggage policies, cross-referenced his calendar, and texted three friends about dates. By the end, he was frustrated and settled for a suboptimal option just to be done.
This isn’t a story about Gen Z’s browsing habits—it’s a story about the fundamental inefficiency of how humans interact with the web. We’ve spent 30 years optimizing for information retrieval when what we actually need is help with getting things done. The browser hasn’t evolved yet from a page viewer or web application launcher to a task executor, leaving billions of people performing digital busywork that software should handle.
We believe agentic browsing represents the most significant shift in human-computer interaction since the iPhone. That’s why Opera is building Neon—our mission to create the first truly agentic browser that doesn’t just show you information, but actively helps you accomplish tasks.
The Efficiency Crisis Hidden in Plain Sight
The average knowledge worker switches between applications 1,100 times per day. They spend 21% of their time searching for information and another 41%-50% on repetitive digital tasks that could be automated.
But the consumer market tells an even more compelling story. Americans spend 7 hours per day on digital devices, with 2.5 hours dedicated to “productivity” tasks—managing finances, shopping, booking travel, researching purchases.
McKinsey estimates that up to 30% of the current US work hours could be automated by 2030 with current technology, representing hundreds of billions in time value annually in the US alone.
The friction isn’t just inefficient—it’s exclusionary. Complex multi-step online processes systematically disadvantage users with disabilities, older adults, and anyone without perfect digital literacy. Agentic browsers democratize digital capability by making complex tasks accessible through natural language.
Why Agentic Browsing Is Finally Possible
Three technical inflection points have converged to make truly agentic browsing viable:
Multimodal AI Understanding: Modern LLMs can interpret both visual layouts and textual content, understanding context across different types of web interfaces. This isn’t just screen scraping—it’s semantic understanding of user interfaces and user intent.
Reliable Action Execution: Browser automation has matured from brittle Selenium scripts to robust frameworks that handle dynamic content, JavaScript-heavy sites, and complex user flows. Combined with AI reasoning, we can now build systems that adapt when websites change.
Economic Viability: The cost of AI inference has dropped 99% in two years while capability has exploded. Running an AI agent that can handle complex browsing tasks now costs pennies, not dollars, per session.
Unlike previous automation attempts (think RPA), Opera Neon – for example – understands context and can handle the unexpected—the CAPTCHA that appears, the popup that blocks content, the form field that moved. It is not following predetermined scripts; It is reasoning through novel situations.
From Human-Driven to Fully Autonomous Browsing
Today’s browsers, albeit powerful, still require humans to do all the heavy lifting. We click, we navigate, we copy-paste between applications like sophisticated digital janitors. Think about the last time you compiled a simple report for work—you probably had one tab open with your project management tool, another with your team’s chat, a third with a spreadsheet, and a fourth with a new document. You, the human, were the only thing connecting them, wrestling disparate information into a coherent summary.
This is analogous to how cars operated for decades. Traditional browsers resemble traditional cars that always require a human behind the wheel, accelerating, braking, and shifting gears. But just as we’re witnessing the emergence of autonomous vehicles, we’re now seeing the first truly agentic browsers that can take the wheel.
With Opera Neon—our first agentic browser launching to the public later this year—we’re demonstrating how humans don’t always need to “hold the steering wheel.”
What Makes a Browser “Agentic”?
Traditional browsers are reactive—they wait for your clicks, your navigation, your manual input. An agentic browser is proactive, designed to understand your intent, assist with tasks, and take actions on your behalf. It is an intelligent collaborator that can work alongside you or independently.
The fundamental difference is execution capability. Where traditional browsers require humans to perform every step of a task, Opera Neon could complete entire workflows autonomously. The browser wouldn’t just show you flight options—it would research, compare, and book your trip. It wouldn’t just display your project management tool—it would analyze tickets, prioritize tasks, and generate reports. It could even create new digital assets like websites, applications, or documents based on natural language descriptions.
The “Jira Test”: An Agent in Action
Let’s say you’re a product manager preparing for a monthly review and you tell your Opera Neon browser: ‘Check our Jira board for any P1 bugs still open for the mobile app release. Cross-reference with our Slack #engineering channel for recent updates, pull the latest user feedback from our support portal, and update the launch readiness document in Google Docs with current status and risk assessment.’
Here’s what happens behind the scenes:
Neon Navigates Across Work Tools: The agent opens Jira to scan active tickets, switches to Slack to parse recent engineering discussions, accesses the customer support dashboard for feedback patterns, and opens the shared Google Doc—eliminating the need to juggle multiple tabs
Neon Synthesizes Cross-Platform Data: Instead of you manually copying bug reports from Jira, searching through Slack threads for engineering estimates, then switching to the support tool to find relevant user complaints, the agent connects all these data points and identifies patterns and dependencies.
Neon Writes the Status Report: The agent updates your launch document with current bug status, engineering confidence levels, customer impact assessment, and recommended go/no-go decision—work that typically requires an hour of tab-switching and manual updates.”
Opera’s Journey to Agentic Browsing
The shift to agentic browsing isn’t incremental improvement—it’s a complete transformation of human-computer interaction. We’re witnessing the same category-defining moment that occurred with the iPhone in 2007 or the first web browsers in the 1990s.
Opera has been pushing the boundaries of browser innovation since the dawn of the internet. Our journey to agentic browsing began in spring 2023 with Opera One, which launched with native integration of AI chat. Throughout 2023-2024, we rolled out contextual AI tools, sidebar integrations, and Aria—our free, native browser AI assistant now used by millions.
But May 2025 marked a turning point with Opera Neon’s limited release—the first agentic browser that can perform useful tasks, not just assist with content.
No more 47-minute booking sessions with 12 open tabs.
Opera Neon is set to launch publicly later this year. Experience the future of browsing and join the waitlist at operaneon.com.
This content originally appeared on Opera News and was authored by Krystian Kolondra
Krystian Kolondra | Sciencx (2025-07-31T12:47:10+00:00) Opera’s vision: agentic browsers will tackle web inefficiency and unlock massive productivity gains. Retrieved from https://www.scien.cx/2025/07/31/operas-vision-agentic-browsers-will-tackle-web-inefficiency-and-unlock-massive-productivity-gains/
Please log in to upload a file.
There are no updates yet.
Click the Upload button above to add an update.