Opera Neon Review: The Confusing Future of AI Browsers

The AI revolution is no longer on the horizon; it has arrived, transforming every facet of our digital lives. At the forefront of this change is the humble web browser, which is rapidly evolving from a simple window to the internet into a proactive, intelligent partner. We're seeing this with Google infusing Gemini into Chrome and startups like Perplexity and The Browser Company reimagining search and interaction. Into this bustling and competitive arena steps Opera with its new offering, Neon.

However, Opera is taking a different path. Instead of offering a free, AI-enhanced experience, it has launched Neon as a premium, subscription-based product costing $19.90 a month. This bold move sets expectations sky-high. Neon promises not just an AI assistant, but a multi-agent ecosystem designed to chat, act, and create on your behalf. But does this ambitious vision justify the price tag? This review dives deep into Opera Neon's features, performance, and overall user experience to determine if it's a glimpse into the future of browsing or a paid beta test that asks too much of its users.

What is Opera Neon? An AI Browser with Three Personalities

To understand Neon, you must first understand that it isn't a single tool; it's a browser built around the co-existence of three distinct AI bots. This multi-agent approach is Neon's core design philosophy, presenting a toggle under the search bar that allows you to switch between a standard internet search and three specialized AIs: Chat, Do, and Make. This structure is both its greatest potential strength and its most significant weakness, offering a centralized hub for AI tasks but creating a confusing user experience where knowing which tool to use is a challenge in itself.

From Ad Blockers to AI Agents: Opera's Browser Evolution

Opera has long been known for integrating innovative features directly into its browser. Neon continues this tradition, built upon the familiar Opera framework that includes a built-in ad blocker, VPN, and a customizable sidebar for apps like WhatsApp and Facebook Messenger. However, with Neon, the AI features are not just add-ons; they are placed front and center on the home and new tab screens, signaling a fundamental shift in the browser's purpose. The browser leverages AI models from both OpenAI and Google, though Opera has not specified which models power which agent. This positions Neon as a dedicated platform for AI-native browsing, a departure from the company's previous focus on privacy and convenience features.

The Crowded Market of AI Browsing

Neon is not launching in a vacuum. It enters an increasingly crowded market where major players and nimble startups are all vying to define the future of AI-powered browsing. Google is integrating its powerful Gemini model into Chrome, promising seamless contextual assistance. Perplexity's Comet browser offers a conversational search experience, while The Browser Company's Dia is also exploring new paradigms. The critical difference is that most of these powerful features are being offered for free. By setting a monthly subscription fee of nearly $20, Opera is making a high-stakes bet that Neon's unique multi-agent system provides enough value to convince users to pay for a product category they have always received for free.

A Deep Dive into Neon's Three AI Agents: Chat, Do, and Make

The entire premise of Opera Neon hinges on its trinity of AI agents. Each is designed for a specific purpose, but their separation and distinct capabilities create a learning curve that can feel counterintuitive. Understanding how each one functions—and fails—is key to grasping the Neon experience.

'Chat': The Verbose and Familiar Assistant

Of the three agents, 'Chat' is the most straightforward and will feel immediately familiar to anyone who has used an AI chatbot. It can be accessed before a search query or from a persistent icon in the top-right corner of the browser, making it readily available for quick research questions. In testing, Chat could handle general queries and answer specific questions about the content of an open web page, such as providing a summary on a complex topic.

However, its performance is far from perfect. The answers provided by Chat were often excessively verbose, forcing the user to sift through dense paragraphs to find a simple piece of information. More concerning was its unreliability. When tasked with summarizing comments from several recent articles, the bot returned a 400-word response explaining that there were no comments to analyze. On another occasion, it incorrectly reported that three articles had zero comments when they actually had four. Instead of admitting it couldn't access the information, it gave the impression that it could and then offered a generic guess about what comments on tech sites typically contain. This tendency to "hallucinate" with confidence undermines the trust essential for a reliable AI assistant.

'Do': The Unreliable Agentic Browser

According to Opera's executive vice president, Krystian Kolondra, the failure to summarize comments was a user error; the 'Chat' bot was the wrong tool for a task that required a click to expand the comments section. For such actions, users are meant to turn to 'Do', Neon's agentic AI designed to take control of the browser and complete multi-step tasks on your behalf. This is where Neon's futuristic promise truly lies: the ability to outsource the mundane clicks and scrolls of web surfing.

In practice, however, using 'Do' is a frustrating and often unnerving experience. When tasked with finding and ordering a floral arrangement, the bot scrolled right past several suitable options and added a "monstrous funerary wreath" to the shopping cart, ignoring user clicks on better choices. There is currently no way to course-correct the 'Do' agent while it is in action, leaving the user as a helpless spectator to its blunders. In another test, it confidently declared a show was sold out for January when a quick manual check revealed many available tickets. This level of unflinching but misplaced confidence makes it nearly impossible to trust the agent with any meaningful task. Furthermore, completing tasks with 'Do' was often slower than simply doing them manually, undermining its core value proposition.

'Make': The Clunky On-Demand App Builder

The third agent, 'Make', is an intriguing concept that allows you to create small, single-purpose web tools on demand. It operates within a virtual computer environment, meaning any software, scripts, or images it uses are sandboxed and don't clutter your personal machine. For example, when asked to create a simple memory matching game with Spanish vocabulary, 'Make' successfully produced a functional, if clunky, game within minutes.

The convenience of this feature is undeniable; once the tab is closed, all the associated files and pictures disappear with it. This offers a clean way to generate temporary tools without long-term digital residue. While the creations themselves may lack polish, the 'Make' agent represents a genuinely innovative use of AI in a browser, pointing towards a future where users can generate bespoke utilities for niche tasks on the fly.

The User Experience: A Frustrating Beta Test

Beyond the core functions of its three agents, the overall experience of using Opera Neon feels less like interacting with a sophisticated piece of technology and more like supervising a "hapless intern". The browser is filled with quirks and bugs that betray its "early access" status, making for a jarring and often unproductive workflow.

The 'Cards' System: Reusable Prompts with Untapped Potential

To help users navigate its AI features, Neon includes 'Cards', which are pre-written prompts that function like "power-ups" for AI interactions. In theory, these could save users time by providing templates for complex or frequently used commands. However, the 'Cards' library currently feels more like a gimmick than a utility. It is largely populated with content from the Neon team, ranging from prompts that rewrite websites in the style of Yoda to more practical news aggregators. Opera's hope is that the platform will eventually fill with useful, user-generated creations, but as of now, there is little of substance available, making it hard to imagine regularly using the feature.

Working with an Unpredictable Assistant

The browser's proactive nature often leads to frustrating interactions. In some instances, an AI system would ask for feedback on a proposed plan and then launch into the task without waiting for a response. This kind of unchecked proactivity is concerning, as it's easy to imagine it going wrong—for instance, sending out LinkedIn connection requests to contacts you only intended to research privately. In an even more bizarre bug, when a user provided positive feedback and told the AI to proceed, it replied, "I'm glad you think so!" and promptly stopped working. While Opera's team states that these are known issues to be fixed, they contribute to an overwhelming sense that Neon is a product that is simply not ready for public use, let alone a paid subscription.

Is Neon Worth the Price? A Competitive Analysis

Ultimately, the value of any product comes down to a simple question: is it worth the cost? For Opera Neon, with its $20 per month subscription, this question is particularly pointed, especially given the state of the market.

Neon vs. Free Alternatives: A Tough Sell at $20/Month

Neon's biggest hurdle is its price. In an era where powerful AI features are being integrated into free browsers like Google Chrome, asking users to pay a premium is a "tough sell". While Neon includes standard Opera features like a VPN and ad blocker, these are not enough to justify the cost when the core AI product is so flawed. Competitors are moving quickly, and the value proposition of a free, seamlessly integrated AI assistant in a browser you already use is incredibly compelling compared to Neon's buggy, confusing, and expensive alternative.

Strengths, Limitations, and Market Position

Despite its many flaws, Neon is not without merit. It offers a tantalizing glimpse into a future where we can outsource the "general mundanity of web surfing" to an intelligent agent. The 'Make' agent's sandboxed environment is a genuinely clever innovation. However, these glimmers of promise are overshadowed by overwhelming limitations. The three-agent system is confusing, the flagship 'Do' agent is unreliable and slow, and the entire product feels like a work in progress that requires the user to adapt to its quirks, rather than a smart browser that adapts to the user. As it stands, Neon is positioned as a tool for early adopters and AI enthusiasts who want to be part of the development journey, not for the average user seeking a reliable productivity tool.

Future Outlook and Broader Implications

Opera has been transparent that Neon is an "early access release" and a "work in progress". This framing, however, clashes with its premium subscription model. It asks users to pay to participate in what is effectively a public beta test. This model may work for niche developer tools, but it is a challenging proposition for a consumer-facing browser.

The experiment also raises important questions about the risks of proactive AI agents. For an AI to take meaningful action on our behalf, we must place an immense amount of trust in it. Neon's performance demonstrates just how far we are from that reality. An agent that hallucinates information, makes incorrect purchases, and acts without clear confirmation poses real risks. The development of agentic AI browsers will require not just technological advancement, but also the creation of robust safeguards and intuitive user controls that are conspicuously absent in Neon's current form.

Conclusion: An Ambitious But Premature Vision

Opera Neon is a bold and ambitious experiment that attempts to push the boundaries of what a browser can be. Its multi-agent system, particularly the agentic 'Do' and creative 'Make' bots, hints at a powerful future for human-computer interaction. However, in its current state, Neon is a product caught between a futuristic vision and a frustrating reality.

The core experience is defined by confusion, unreliability, and a steep learning curve. The separation of Chat, Do, and Make is not intuitive, and the performance of each agent ranges from verbose to dangerously inaccurate. When you combine these fundamental flaws with a $20 monthly subscription in a market of powerful free alternatives, the value proposition collapses. Opera Neon feels less like a finished product and more like a browser we are expected to adapt to, rather than one that is smart enough to adapt to us. For now, it remains a tough sell and a cautionary tale in the exciting but challenging race to build the true AI browser.

Frequently Asked Questions (FAQ)

1. What are the three main AI features in Opera Neon?

Opera Neon is built around three distinct AI agents: 'Chat', a conversational assistant for answering questions; 'Do', an agentic AI that can take actions and control the browser to complete tasks; and 'Make', a tool that can build simple, temporary web applications on demand.

2. How does Opera Neon's "Do" agent compare to a standard AI chatbot?

Unlike a standard AI chatbot like 'Chat' which only provides information, the 'Do' agent is an "agentic" AI. It is designed to take over the browser to perform actions for you, such as booking appointments or adding items to a shopping cart, though its current performance is unreliable.

3. Why is Opera charging for Neon when browsers like Chrome have free AI?

Opera is positioning Neon as a premium product with a unique multi-agent system not found in competitors. The $19.90 monthly fee is a bet that early adopters and AI enthusiasts will pay for access to this experimental platform, even while competitors like Google integrate AI into their browsers for free.

4. What are the biggest limitations of Opera Neon in its current state?

The primary limitations are its confusing three-agent structure, the unreliability and slowness of the 'Do' agent, the verbosity and inaccuracy of the 'Chat' bot, and a variety of bugs that make it feel like an unfinished product. The high price is also a significant barrier for most users.

5. Is Opera Neon suitable for everyday use?

In its current "early access" stage, Neon is not suitable for most users as a primary, everyday browser. It is better suited for AI enthusiasts and developers who are interested in experimenting with new browser concepts and are willing to tolerate significant bugs and performance issues.

6. What happens when the 'Do' agent in Neon gets stuck?

When the 'Do' agent encounters an obstacle it cannot handle on its own, the 'Do' tab at the top of the screen will flash red. This is a signal that the user needs to intervene and provide input to help the bot continue with its task, highlighting that it is not yet a fully autonomous system.