AI Browsers Are Here, What Atlas and Comet Mean for the Future of Marketing

Over the past few years, AI has gone from a tool used by experimental marketers, looking for a new take on their work, to an everyday resource changing the way we act online. 

From OpenAI to Perplexity, AI companies are pumping out system updates, exciting features and new platforms at a speed that would’ve been unfathomable in the past. 

Some of the biggest updates we’ve seen include the launches of OpenAI’s Atlas and Perplexity’s Comet. AI search engines are set to not only change how we market on search, but also how users ultimately behave online. 

For marketers, it’s imperative to start understanding the impact of these changes as soon as possible. Waiting for AI search engines to become the new normal will risk a loss of visibility, data accuracy and how our brands are represented.

In this blog, we’re going to deep dive into the world of AI search engines, looking at its potential impacts in buyer behaviour and what this means for the future of marketing. 

So, let’s jump straight in.

What Are AI Browsers? How Atlas and Comet Work

In simple terms, AI browsers are search engines, hosted by LLM’s (large language models) which utilise autonomous agents to enhance the overall user experience. Rather than responding to general search queries, they instantly source information for the user without them having to click into a website, similar to Google’s AI Overview. 

However, AI browsers can take this function a step further, by interpreting instructions, offering opinions and even completing purchases and bookings through agentic ecommerce.

Some of their key features include: 

Conversational search

• Users will use conversational language, such as ‘where is the best coffee shop near me?’ and AI will produce a list of local venues that it assumes the user will like. This can be expanded on, for example, ‘which of these has the best atmosphere?’, and it will delve into the key features of each venue, offering a solution that the user may prefer.

Personalised experiences

• If users already have accounts with OpenAI or Perplexity, chat history will be transferred over to the new browsers upon log ins, creating an instantly personalised experience. 

Task completion inside the browser

• Agentic ecommerce is where, upon instruction from the user, AI is to complete online tasks on their behalf. Like, buying a new pair of shoes or booking a yoga class.

Ability to interpret spreadsheets and structured data

• AI browsers will be able to interpret data from online spreadsheets, on applications like Google Sheets and SharePoint. They’ll be able to dissect a spreadsheet and give you a detailed summary of its findings in a matter of seconds. 

But why does this matter to marketers?

These platforms are purposely built to offer users an online experience that they have never encountered before and to ultimately change the way we use search. 

For marketers, this means we must change the way we strategise to ensure our brands aren’t losing visibility.

 

How AI Browsers Change the Search Experience

The most important impact of AI browsers is how dramatically they reduce friction in the search journey. 

Traditional Search Behaviour

Historically, search followed a more predictable pattern. Users would input their search request into google, review the results and click through multiple websites until they found the information they needed. 

This worked well for websites with a strong SEO presence that had strong SERP results and features based on EEAT. 

Visibility and performance were very closely tied to traffic. Each visit meant multiple opportunities for brands to communicate their value through engaging copy and captivating landing page designs. With the aim of guiding users through a journey and ultimately leading them to convert. More traffic generally meant more chances for the site to perform. 

Performance metrics reinforced this relationship, measuring success through: 

• Organic sessions 
• Click-through rates
• Time on site 
• Conversions attributed to visits

AI-Driven Behaviour

AI browsers are built to optimise this journey, compressing the entire process to be as efficient for the user as possible. 

By using questions or instructions as prompts, the AI agent crawls for information, determines the best course of action, and completes it – often without showing the user underlying sources. 

This results in brand’s drastically losing traffic, but users having a much shorter purchase journey. 

The browser becomes both the search engine and the action layer. 

As marketers, the key implication is that we’re no longer optimising for clicks. We’re optimising for inclusion, accuracy, and trust within AI outputs.

 

Agentic Ecommerce and Agent Mode Explained

One of the most disruptive aspects of AI browsers is the introduction of agent mode. 

What does agent mode mean?

Agent mode allows the browser to act autonomously if permission is granted by the user. Instead of assisting them through each step, the agent completes tasks independently. 

This includes multi-step processes that would normally be completed by a human. This is a complete shift in how we as humans behave in the online world.

What Agents can do today:

In agent mode, AI browsers can:

• Make purchases
• Book appointments
• Complete checkout flows 
• Compare products and prices 
• Carry out repetitive administrative tasks, like booking appointments

This fundamentally changes ecommerce, customer journeys, and conversion paths. 

The ‘buyer’ is no longer required to be human. 

Shopify was one of the first platforms to announce its strategic partnership with OpenAI, enabling users to purchase through websites they host, utilising agentic ecommerce.

Security and Reliability Risks 

Current research and early implementations highlight several risks that marketers and businesses must understand. 

Including: 

• AI agents acting on inaccurate or incomplete information 
• Vulnerabilities such as prompt injection/manipulation 
• Exposure of sensitive connected data and accounts 
• Theoretical potential for coordinated malicious use

A critical issue is that agents will execute instructions; they do not have ‘better judgment’. If the input or source material is flawed, the outcome can still be completed successfully. 

 

The Risks for Brands and Marketers Using AI Browsers

Declining Website Visibility

AI Browsers are designed to keep users inside their platform. 

They encourage conversation and try to offer the user the answer in one place, again ensuring that the user journey is drastically shortened. 

Continuing to reduce the need for users to visit external websites, resulting in fewer website visits and brand impressions.

Over time, this will lead to the redundancy of content that was designed to attract traffic and increase referral traffic tenfold. 

Loss of Control Over Brand Representation

AI browsers rarely quote content verbatim, instead they interpret and summarise. 

This introduces a loss of control over:

• Messaging 
• Tone 
• Content 
• Accuracy

Brands may be represented by partial information, outdated sources or simplified summaries with no visibility into how or why that content was chosen by the agent. 

 

Atlas Does Not Validate Accuracy

Atlas, like many AI agents, does not verify outcomes independently.

It relies entirely on the instructions provided by the user, retrieved sources and model interpretations. Whilst this can provide an efficient search experience, it can result in: 

• Incorrect actions 
• Inconsistent outputs
• Errors that appear ‘successful’ from a system perspective

For brands, this creates reputational risk if AI-generated actions or summaries are incorrect.

 

Privacy and Trust Concerns

To work as effectively as they advertise, AI browsers need to access an excess of user data to create an optimised online experience. 

But, with the goal posts in AI changing so abruptly, there is no clear information on how data is found, used and stored. 

Some of the data they need to access includes: 

• Browsing history 
• Emails & calendars
• Payment methods
• Connected SaaS platforms

For users this raises trust concerns around their secure data and for brands it introduces questions around compliance, data handling and indirect exposure. 

 

Advertising and Analytics Problems Created by AI Browsing Tools

Ad Interactions That Look Human

One of the immediate issues for people in the marketing sphere is how AI browsers interact with advertising. 

Atlas can click ads, navigate landing pages, and trigger engagement events in ways that closely resemble human behaviour. 

Atlas runs on Chrome, which means ad networks and analytics tools recognise the actions as if they were being completed by humans. There is currently no tool that can tell them apart. 

Unfortunately, this means that ad platforms are unable to tell the difference between paying for AI agents or humans. Which could lead to ad budgets being drained by AI crawlers, which never had the intent to convert. 

 

Rising Risk of Ad Fraud

As AI-generated activity becomes more sophisticated, distinguishing it from real users becomes increasingly difficult. 

This creates conditions where: 

• Spend can be inflated unintentionally 
• Campaign performance appears stronger than in reality 
• Coordinated misuse become possible

This opens the floodgates for a new security risk, where an AI agent could unintentionally click on a malicious link without the user’s awareness. 

With agentic ecommerce still being quite the unknown, it’s also possible that we see a rise in coordinated sabotage or inflated spend. As unethical brands could target their competitors with ‘spam’ crawlers to click through to their website, mimicking human sessions. 

Unfortunately, traditional fraud detection systems are currently unable to find this type of behaviour.

 

Analytics Pollution

AI traffic blends into analytics platforms with little to no differentiation. 

These make current attribution models unreliable; conversion rates will decline without a clear cause, and optimisation decisions will be based on distorted data.

Over time, this erodes confidence in performance reporting and makes it harder to justify budgets or strategic changes. 

 

Future Predictions: The Impact of AI Browsers on SEO and PPC

Based on current market signals, several trends are becoming increasingly more likely to pop up in the marketing world.

First, it is highly likely that Google will release its own AI-native browser. Google already holds a monopoly over the online world, so making this move would be a logical decision for them. Especially before OpenAI or Perplexity start pushing users to make Atlas or Comet their default browsers. 

If AI browsers reach mainstream usage: 

• SEO will shift toward AEO (answer generative optimisation) 
• PPC measurement will become less reliable 
• Visibility will depend on inclusion, not ranking 
• Agencies will need to redefine value beyond traffic acquisition 

Content strategies will increasingly focus on being interpretable, trustworthy, and contextually rich rather than purely optimised for clicks. 

 

What Marketers Can Do Now to Prepare for AI-Driven Search

Audit Your AI Presence

Marketers should actively test how their brand is appearing in AI browsers, conversational search tools and agent driven recommendations. 

This provides early insight into gaps, inaccuracies and any missed opportunities.

 

Monitor Analytics for Irregular Patterns

AI activity often presents as subtle anomalies rather than obvious spikes, so it’s important to look out for key signals that may indicate non-human behaviour.

Including: 

• Sudden increases in traffic with no campaign changes
• Unusual click-through rates 
• Declining conversion rates despite strong engagement 

Create AI-Ready Content (AEO)

Content designed for AI interpretation should be: 

• Highly structured
• Context-rich
• Clear and unambiguous 
• Actionable and specific 

This improves the likelihood that summaries are accurate and useful, rather than generic or misleading.

Stay Educated as Standards Evolve

AI browsing standards are not yet fixed, so it’s imperative that, as marketers, we stay up to date on important updates that could completely reframe our marketing strategies. 

You can do this by subscribing to marketing newsletters like Search Engine Land or keep an eye out for press releases from OpenAI, Perplexity, Microsoft and even Google.

Early understanding of these updates creates a strategic advantage.

 

Opportunities: How AI Browsers Support Automation for Marketers

Despite the risks and the overall concerned feeling that surrounds AI browsers, they could bring meaningful opportunities to the table for both brands and users. 

When used intentionally, they not only offer the user an enhanced online experience, but they can also offer brands: 

• Faster research and competitor analysis 
• Automated reporting and data interpretation
• Rapid extraction of insights from large datasets

For internal workflows, AI-powered browsing tools can significantly reduce time spent on manual analysis and repetitive tasks. 

The key is controlled and intentional use, no blind reliance. 

 

Conclusion

AI browsers represent one of the most significant shifts in digital behaviour since the rise of search engines themselves. 

They reduce friction for users, automate decisions and compress journeys. But they also disrupt visibility, attribution and control for brands. 

For marketers, waiting for these shifts to take place is risky. 

Understanding how these systems work, how they represent brands, and how they interact with advertising is now a strategic necessity. 

The future of marketing will not only be defined by who earns the click, but by who is understood, trusted, and correctly represented by AI.

Adaptation is no longer optional; it is the next phase of competitive advantage for brands.