How Will AI Web Browsers Change The Way We Browse?

The way we interact with the internet is on the cusp of an evolutionary jump, driven by the emergence of AI-powered web browsers. No longer content with merely displaying information, these next-generation browsers, exemplified by Perplexity’s recently launched Comet and OpenAI's soon-to-be-revealed AI browser, are poised to redefine how we interact with the internet. This shift promises new levels of speed and personalisation, but it also casts a long shadow, particularly over the future of Search Engine Optimisation (SEO) and, more critically, the potential for digital echo chambers.

‍

The Dawn of the AI Web Browser

Traditional web browsing involves a manual journey: typing queries, sifting through search results, clicking on various links, and synthesising information across multiple tabs. AI web browsers aim to replace this process with a seamless, intuitive experience. Imagine a browser that replaces your search results with direct, summarised answers instead of links. Sure, we are seeing something similar with Google’s introduction of AI Overviews (AIOs), but AI browsers from Perplexity and OpenAI intend on taking this a step further.

Integral to AI Browsers is a native AI assistant that can deliver personalised conversational interactions with users and carry out a range of actions on their behalf, including sending emails, booking meetings, checking items in stock across different stores, summarising videos or articles, and much more. Assistants within AI browsers are context-aware and able to reference any open tabs to perform tasks without you having to copy and paste content into the conversational interface.

Perplexity announced Comet in early July 2025, to “power a shift from browsing to thinking”. It replaces search with Perplexity native in-browser, and can complete a range of tasks on behalf of users via the Comet Assistant.

Just a few days following Comet’s release, OpenAI teased the release of their own AI Browser, expected to be released in the coming weeks. OpenAI’s browser is expected to follow a similar path, leveraging its powerful AI model to offer proactive, agent-led navigation.

This marks a drastic shift from "Browse" to "cognition". The browser doesn't just show you the internet; it actively helps you think, research, and act.

‍

How will AI Browsers Change Our Internet Experience?

The immediate impact on how humans access the internet will be greater levels of personalisation and an increase in the speed we can find information and carry out day-to-day tasks. Here’s a few ways that AI browsers are poised to change how we use the internet:

• Direct Answers and Task Automation: Need to find the best local coffee shop, compare prices on a new gadget, or schedule a meeting? A native assistant within an AI browser could handle the process with minimal user input. Imagine you’re trying to buy the new Nintendo Switch 2, you can ask the AI to look for local stores in your area as well as ecommerce retailers in your region and present any back to you that have stock on hand. For anyone who endured the stock shortage of the PS5 launch, this is far easier than crawling through websites and waiting on Tweets for new stock releases.
• Information Synthesis: For researchers, students, or anyone dealing with information overload, AI browsers can save hours in research. They can digest vast amounts of data from multiple sources and present concise, coherent summaries, highlighting key facts and insights.
• Hyper-Personalisation: The browser will learn your preferences, habits, and even your unique conversational style, tailoring content and recommendations with unprecedented accuracy. This promises a more intuitive and relevant online experience.
• Enhanced Accessibility: AI features like real-time translation, voice commands, and content simplification could make the internet far more accessible for individuals with disabilities, breaking down traditional barriers to information.

‍

Is Privacy a Concern?

It seems that with every technological convenience that we embrace with open arms, we knowingly or unknowingly embrace privacy issues simultaneously. With Perplexity Comet, you do get a native ad blocker out-of-the-box, which is nice, but according to their privacy policy, "Your input and output, such as questions, prompts and other content that you input, upload or submit to the Services, and the output that you create, and any collections or pages that you generate using the Services" will be retained by Perplexity.

Additionally, if you connect Perplexity to your Google account so that it can perform tasks like sending emails and setting up meetings, you’ll also be feeding this data to Perplexity.

‍

Could personalisation lead to misinformation and echo chambers?

You only have to spend 5 minutes online to gauge the sheer volume of misinformation that exists on the internet. From unsubstantiated facts to satirical news websites that some less-savvy users believe to be true, and sarcastic comments on otherwise helpful Reddit threads.

One concerning implication on a more personalised web browsing experience via AI browsers is the potential for creating “echo chambers” and amplifying content based on user preferences. We’ve seen this effect play out in social media feeds, as they prefer content that users are more likely to engage with. As AI browsers begin to understand the user and their preferences, could this lead to a similar effect?

Here's why it could be a concern:

• Algorithmic Reinforcement: AI constantly learns from your interactions. If you tend to agree with a certain political viewpoint or prefer a particular type of news, the AI will prioritise showing you more of that content. This reinforces existing beliefs, creating a feedback loop where you're primarily exposed to information that confirms your biases.
• Lack of Diverse Perspectives: The inherent goal of personalisation is to filter out what you don't want. While this is convenient, it can inadvertently filter out dissenting opinions, alternative viewpoints, and information that challenges your worldview. This leads to a narrow, curated understanding of complex issues.
• Erosion of Critical Thinking: If you are constantly fed information that aligns with your beliefs, there's less incentive to critically evaluate different arguments or challenge your own assumptions. This can hinder intellectual growth and lead to a less nuanced understanding of the world.
• Amplified Misinformation: Within a closed echo chamber, false or misleading information can spread unchecked, as there are fewer counter-narratives or critical voices to challenge its veracity.

‍

How are Perplexity and OpenAI combating this?

Recognising this significant ethical challenge, both Perplexity and OpenAI have indicated approaches to mitigate the echo chamber effect, though the long-term effectiveness remains to be seen.

Perplexity, with its new Comet browser, leans heavily on “source transparency and accuracy”. Its core value proposition is to provide direct answers with prominent citations, empowering users to:

• Verify Information: Users can easily click through to the original web sources, enabling them to assess the credibility and context of the information themselves.
• Explore Diverse Viewpoints: By listing multiple sources, Perplexity implicitly encourages users to explore the origins of different claims. Furthermore, Perplexity Max subscribers can utilise features like "Ask Comet for counterpoints" to explicitly request alternative perspectives on a topic.
• Prioritise Authoritative Sources: Perplexity's algorithms aim to de-emphasise low-quality or highly biased content, striving to present information from reputable outlets.

OpenAI, with its extensive research into “Responsible AI”, is expected to embed bias mitigation into the very core of its browser's design. Their strategies likely include:

• Alignment and RLHF: OpenAI's ongoing research into "alignment" aims to ensure AI systems act in accordance with human values, which includes efforts to reduce harmful biases through techniques like Reinforcement Learning from Human Feedback (RLHF).
• Mitigation in Training Data: They invest heavily in curating and filtering vast training datasets to reduce inherent biases found in internet content.
• Transparency & Limitations: OpenAI typically discloses the capabilities and limitations of their models, aiming to make users aware that AI can generate incorrect or biased information, encouraging user diligence.
• User Feedback: Like ChatGPT, their browser would likely include robust feedback mechanisms, allowing users to flag biased or inaccurate responses, which can then be used to refine the AI.

Despite these efforts, the fundamental tension between personalisation (giving users what they want) and objectivity (presenting a balanced truth) will persist. It requires a conscious effort from both AI developers in their design principles and users in their approach to information consumption.

‍

How will this impact Google and SEO?

Comet’s announcement came with a jab from Perplexity on their “slow” speed to innovate in AI. Google Chrome, with a global market share of approximately 68% undoubtedly has the most to lose from new competitors. While a new alternative web browser is an occurrence Chrome has weathered many times before, AI browsers like Comet also seek to divert traffic from Google’s bread and butter, Google Search.

With Comet replacing search with its own results, users won’t see Google Search results unless they specifically navigate to Google. This poses an existential question for Search Engine Optimisation (SEO) professionals: how can they reach searchers using Comet or another AI browser instead of Google? While this isn’t necessarily a new question, as SEO professionals have started focusing more heavily on optimising content to be found and sourced by AI models, it does amplify the importance of this.

Luckily for SEO professionals who are keeping up with current demands, the space was already headed in this direction, with less of a one-dimensional focus on keywords towards a more robust focus on aligning with search intent and building authority through expertise and experience.

The real impact could be on businesses who rely heavily on Organic Google Search as a primary source of traffic. If AI provides an answer and even navigates websites to complete tasks for you, what will the financial impact be to these businesses?

It is worth noting that if AI browsers like Comet do syphon a decent chunk of Google Chrome’s market share, Google is still firmly the lead in search, processing 13.7 billion searches per day, which is approximately 373 times the estimated 37.5 million daily search-like prompts that ChatGPT receives. Tim will tell if AI browsers moves the needle for Google.

‍

Navigating the AI-Powered Internet

For us, the internet users, this new era demands heightened digital literacy and critical thinking. We must understand that convenience can come at the cost of breadth, and personalised bubbles can distort reality. It's crucial to:

• Actively Seek Diverse Information: Don't rely solely on what your AI browser presents. Deliberately seek out different news sources, opinions, and perspectives.
• Verify Information: Always utilise the provided citations to check original sources and assess their credibility.
• Understand How AI Works: Remember that AI is a tool, not an oracle. It can make mistakes, and its knowledge is always based on its training data.
• Demand Transparency and Control: As users, we should advocate for AI browser developers to be transparent about their filtering mechanisms and provide granular controls over personalisation settings.

AI browsers could represent a huge shift in how we use the Internet. While there are some elements to be wary of, we’re excited to see where this takes us.

‍