I still remember the first time I heard about a browser that could actually do things for me, not just show me web pages. The idea seemed almost absurd - like handing your car keys to an algorithm and hoping it knows where you want to go. Yet here we are in October 2025, and Opera has just launched something that makes that metaphor feel uncomfortably accurate.
Opera Neon isn't your typical browser update with a fresh coat of paint and some tweaked features. This is Opera resurrecting a brand they experimented with back in 2016-2017, but now turbocharged with artificial intelligence that doesn't just answer questions - it acts on them. The company calls it an "agentic browser," a term that sounds like corporate jargon until you see what it actually means in practice.
The browser officially shipped on September 30, 2025, though getting your hands on it requires jumping through hoops. Opera rolled out invites progressively, creating a Founders' Program for early adopters willing to pay $59.90 for nine months of access. After that promotional window closes, the price jumps to $19.90 monthly. In a world where Chrome and Edge cost nothing, that's a bold move - some might say audacious, others might call it delusional.
The Architecture of Action
What separates Neon from the dozens of browsers with chatbots slapped onto their sidebars? The answer lies in how Opera structured the entire experience around three distinct operational modes that fundamentally change your relationship with web browsing.
Chat mode handles the familiar territory - conversational AI that can answer queries, summarize pages, and pull contextual information from your browsing history. It's competent but unremarkable in 2025's crowded AI landscape. Everyone has a chatbot now. Even your refrigerator probably has one.
Make mode ventures into more interesting territory, letting you generate content ranging from code snippets to full applications, images, and since October 7, 2025, even videos through integrated OpenAI Sora 2 support. Type "@make [your prompt] using Sora 2" into the Omnibox, and Neon creates video content with shareable links hosted on operaneon.com. For content creators juggling multiple AI subscriptions, consolidating video generation directly into the browser offers genuine convenience.
But Do mode - that's where Opera either proves their vision or reveals their hubris. This is the agentic core, the feature that justifies the "built to act" tagline plastered across their marketing materials. Do mode doesn't wait for you to manually compare prices across ten tabs or copy-paste data between spreadsheets. It opens websites, navigates pages, evaluates security risks, fills forms, and even executes transactions while you watch in real-time, pausing for confirmation on sensitive actions.
The system operates through Tasks - isolated workspaces that maintain complete context across multiple tabs, documents, and AI interactions without data bleeding between sessions. Planning a vacation? Neon can simultaneously analyze flight options, compare hotel prices, check reviews, calculate budgets, and present organized recommendations. Each Task functions as a contained environment where the AI understands relationships between different information sources and can act across them coherently.
Cards extend this automation framework through reusable prompt templates. Think of them as macros for the AI age. You can create a Card for "weekly meal planning with shopping list generation" or "competitive price monitoring for specific products," then trigger these complex workflows with single commands. Opera built a community marketplace where users share Cards, similar to how IFTTT popularized automation recipes. The utility depends entirely on whether this ecosystem develops genuine depth or becomes cluttered with low-quality templates.
The Privacy Paradox
Here's where Opera makes claims that sound almost too good to be true in 2025's surveillance-heavy internet. Neon processes operations locally within your browser session wherever technically possible, avoiding cloud authentication for sites where you're already logged in. The AI doesn't need your passwords to act on your behalf - it uses your existing browser sessions.
This local-first architecture addresses legitimate concerns. When AI agents need access to your banking, shopping, or email accounts to perform tasks, you want assurances they're not transmitting credentials to remote servers. Neon inherits Opera's legacy privacy features - tracker protection - which performed respectably in tests against tools like Cover Your Tracks.
But "wherever possible" carries weight that deserves scrutiny. Some AI operations inevitably require cloud processing. The browser integrates models from OpenAI (including GPT and Sora 2) and Google (for image generation and other capabilities). Understanding exactly what data gets transmitted when, and to which providers, remains crucial for users who actually care about privacy beyond marketing promises.
The model-agnostic approach means Neon can theoretically tap whatever AI capabilities work best for specific tasks, but it also means multiple potential data pipelines to external services. Opera emphasizes user visibility - you can see what the AI is doing in real-time and intervene at any point - but transparency about actions differs from transparency about data flows.
The Economics of Automation
Let's address the elephant in the browser: that $19.90 monthly subscription. Opera positions this as premium pricing for power users who value time over money, professionals handling complex workflows where automation delivers measurable efficiency gains. The math only works if you genuinely save hours monthly through automated tasks.
The early bird offer - $59.90 for nine months - works out to roughly $6.66 monthly, which sounds far more palatable. This pricing strategy clearly aims to build an initial user base willing to evangelize the product while Opera refines the experience based on feedback. The Founders' Program creates artificial scarcity and exclusivity, psychological tactics that can backfire if the product doesn't deliver proportional value.
Compare this to the competitive landscape. Perplexity recently made Comet free with an optional $5 monthly plan for enhanced news features. Microsoft bundles Edge Copilot at no extra cost for Windows users. Google Chrome's forthcoming AI capabilities will likely follow their standard model - free for users, monetized through data and advertising. Arc from The Browser Company offers innovative features without subscription fees.
Opera's betting that deeper, more reliable automation justifies premium pricing. But the market has shown repeatedly that consumers resist paying for software when functional free alternatives exist. The question isn't whether Neon offers $20 of value monthly. It's whether enough users recognize that value quickly enough before competitors close the capability gap.
Reality Testing
Early adopters report mixed experiences that paint a more nuanced picture than Opera's polished marketing materials suggest. The interface looks familiar - essentially Opera's established design with AI integration woven throughout. A chat sidebar, prominent AI button, and new tab page featuring quick access to agentic tools. Nothing groundbreaking visually, which might actually be wise. Radical interface changes often alienate existing users.
Make mode impressed testers by generating functional retro games complete with working code and assets in minutes. That's legitimately useful for developers prototyping concepts or non-programmers experimenting with interactive content. The Sora 2 integration for video generation works smoothly according to initial reports, with users sharing positive experiences about creating shareable video content directly in the browser.
But Do mode's agentic capabilities show inconsistent reliability in complex scenarios. The AI sometimes misinterprets ambiguous instructions or struggles with websites using unusual navigation patterns. One reviewer noted that while simple tasks like price comparison work well, complicated multi-step workflows occasionally fail in unpredictable ways. The system pauses for user confirmation on sensitive actions, which provides safety but interrupts the automation that's supposed to save time.
Chat responses occasionally lack direct source links, making verification harder than it should be. This matters when you're relying on AI-gathered information for important decisions. The promised offline functionality remains unclear in actual implementation - some sources mention AI agents continuing work when users are offline, but practical demonstrations of this capability are scarce.
Security researchers have raised concerns about agentic browsers generally. Similar tools have been tricked into performing unintended actions through carefully crafted web content. If an AI agent can navigate websites and execute actions on your behalf, what happens when it encounters a malicious site designed to exploit that automation? Opera will need to demonstrate robust safeguards beyond what they've shown so far.
The Competitive Battlefield
Neon enters a crowded and rapidly evolving market. Microsoft's Edge Copilot, while less sophisticated in agentic capabilities, comes bundled with Windows at no extra cost. Google's Project Mariner represents their exploration of similar territory, backed by resources Opera can't match. The Browser Company is developing Dia with their characteristic design-forward approach.
Opera differentiates through the Tasks system's contextual isolation and Cards marketplace for shared automation templates. These architectural choices matter if you're the target user handling genuinely complex workflows. But most people don't. Most people browse social media, check email, shop occasionally, and watch videos. For those use cases, free browsers with basic AI assistance suffice perfectly.
The technical implementation shows Opera learned from their decades in the browser business. Built on modern web standards, supporting Windows and macOS (though notably absent on mobile platforms where most browsing now happens), and maintaining compatibility with existing web technologies. They're not reinventing the wheel - they're adding an AI engine to an established vehicle.
What's genuinely interesting is Opera's broader strategy. They're simultaneously upgrading their free flagship browsers with faster AI models and local open-source options while positioning Neon as the premium tier for advanced automation. This creates a product ladder where casual users get free AI features, while professionals willing to pay access deeper capabilities. Whether these products complement or cannibalize each other will determine if the strategy succeeds.
Who Actually Benefits
Let me be specific about who might genuinely find $240 annually justifiable. Academic researchers comparing information across dozens of sources, synthesizing findings, and managing citation workflows could reclaim significant time. Journalists investigating stories that require monitoring multiple sources and tracking developing narratives might automate substantial portions of their information gathering.
Competitive intelligence analysts who spend entire days comparing competitor websites, pricing, features, and market positioning could use Tasks to systematize what's currently manual drudgery. Financial professionals monitoring market conditions across multiple platforms might automate data collection and initial analysis.
Small business owners juggling vendor relationships, product sourcing, and administrative tasks could create Cards for recurring workflows - automated quote requests, price monitoring, invoice processing. The ROI calculation becomes straightforward if you're billing $100+ hourly and Neon saves you five hours monthly.
Content creators needing frequent video assets or image generation might justify the cost through consolidated tooling. If you're currently paying for multiple AI services, consolidating some functionality into your browser while maintaining better privacy could make financial sense.
But - and this matters - casual users gain almost nothing worth paying for. If your browsing consists mainly of consumption rather than complex information work, free alternatives handle your needs perfectly. The subscription model inherently limits Neon's potential market to a small percentage of internet users.
The Honest Calculation
After examining everything Opera promises and what early users actually experience, I find myself skeptically curious rather than convinced. The technical approach appears sound. Local-first processing addresses real privacy concerns. The Tasks and Cards architecture offers organizational benefits for specific workflows. Integration with cutting-edge AI models provides capabilities competitors lack today.
But several factors give me pause. First, the platform limitation to desktop computers in an increasingly mobile-first world significantly constrains the addressable market. Second, the invite-only launch creates artificial scarcity that might backfire if growth stays too slow. Third, and most critically, that monthly subscription faces competition from free alternatives improving rapidly.
Opera is essentially betting that a meaningful percentage of users will recognize enough value in superior automation to justify ongoing payment. That's a tough sell historically. Software subscriptions succeed when they're either essential infrastructure (email, collaboration tools) or deliver continuous value that free alternatives can't match. Browsers traditionally fall into neither category.
The Cards ecosystem needs critical mass to become genuinely useful. If adoption stays limited, the marketplace won't develop the depth required to deliver on its promise. It's a chicken-and-egg problem - users need compelling Cards to see value, but compelling Cards require a large user base creating and refining them.
Maybe Opera proves everyone wrong. Maybe enough professionals recognize that automated browsing saves them hours weekly, justifying the cost easily. Maybe the Founders' Program creates a passionate early community that evangelizes the product and builds amazing Cards that demonstrate undeniable value.
Or maybe this becomes a cautionary tale about charging for capabilities that everyone else will offer free within months. About launching desktop-only in a mobile world. About betting on agentic AI before the technology proves reliable enough for mainstream users to trust.
What's undeniable is that we're watching browsers transform from passive windows into active participants in our digital lives. Whether that transformation happens through paid subscriptions or free integrated features remains uncertain. What's clear is Opera is making a significant bet on the former, and the next year will reveal whether they read the market correctly or miscalculated badly.
The browser that acts for you is here. The question is whether you'll want it badly enough to pay for it.