A new document from the publicly traded market research and intelligence agency SimilarWeb, which covers international web traffic patterns for AI-related platforms for 12 weeks through May 9, 2025, provides a useful look for businesses and serious users into the present landscape of relational AI usage online.

The report tracks changes across various industries, including general-purpose AI equipment, coding assistants, content creators, and more, using custom analytics based on page visits.

Additionally, it depicts the effects of conventional industries like knowledge, search, and online freelancing.

Why are enterprise professional decision-makers interested in this report?

This review is more than just a client trend snapshot for enterprise AI leaders, especially those in charge of model deployment, orchestration, or data integration. It’s a chart of what users are used to and expect.

People are already using tools like ChatGPT, Claude, DeepSeek, and Grok at household or freely experiment with them at work. If your domestic AI apps or copilots don’t suit the standard experience provided by these tools, you run the risk of users being rejected and not adopted.

Additionally, choosing the exact models with high outside grip for your inner orchestration pipelines or RAG deployments can save time and money on onboarding because your team will already be familiar with the interface and behavior.

In summary, the quickest way to respect, usability, and value delivery may be to align your enterprise load with the tools dominating usage charts.

Here are the report’s five main conclusions:

Adoption for developer-focused AI tools is on the rise, with customers to the category increasing by 75 % in the last 12 months.

Lovable, which experienced a jaw-dropping + 17,600 % increase, and Cursor, which experienced steady month-over-month growth.

Use trends suggest that these tools are no longer only experimental tools; they are now seen as essential equipment for modern technology teams as AI becomes more deeply embedded in IDEs and constant connectivity workflows.

As the use of these tools increases, OpenAI wants to take a bigger share of the activity and revenue, particularly since its models are frequently used as the website behind-the-scenes for these very users, especially given that OpenAI apparently tried to buy Cursor and has also officially finalized a deal to buy rival AI coding platform Windsurf.

2. We all know that Grok and DeepSeek had a moment earlier this year, but we all know that both have since fallen back into low plateaus.

Grok and DeepSeek, two of the AI platforms that are growing the quickest this year, demonstrate how quickly hype can turn into burnout.

Grok traffic increased by more than 1, 000, 000 % in March, as a result of its reputation as an open yet illuminating platform and Elon Musk association, before falling by more than 5, 200 % by early May.

Similar arcs were seen in DeepSeek, peaking at +17 and registering 70 % growth before falling to -41 %. Takeaway: Despite OpenAI’s and Google’s legacy tech brand’s, virality can’t replace retention.

Writing and content tools are experiencing user fatigue, once one of the most widely used use cases for generative AI.

Overall, category traffic decreased by 11 %, with platforms like Wordtune ( 35 % ), Jasper ( 19 % ), and Rytr ( 23 % ), all trending downward.

Only originality, please. Ai reversed the trend by consistently increasing traffic, likely as a result of its emphasis on AI generation rather than detection. This plateau points to increased content saturation and potential for improvement in skepticism or usefulness.

In contrast to the SimilarWeb report, I believe that the majority of users who are interested in chatbots will instead use a model-building tool like OpenAI’s website for ChatGPT ( and mobile apps ), Google’s Gemini, or even Anthropic’s Claude directly because they already have a default text input/output field as their main user interface, even for multimodal interactions.

AI with a design focus is still a mixed bag. While overall category usage decreased marginally (-6 % over the 12-week window ), some platforms experienced significant increases.

Artbreeder increased by + 100 %, while Getimg posted a significant + 1, 532 % surge.

Other companies experienced double-digit declines, such as Looka and Stable Diffusion. The odd pattern may be a result of a crowded field of tools that provide comparable functionality but compete for aesthetics or novelty.

5. AI is destroying traditional methods of technology, including crowdsourcing and search.

Traditional digital services are showing general, slow declines, most likely the result of AI substitution.

Users turning to AI tools for tasks like design, writing, and code, may be causing traffic losses on freelance platforms like Fiverr (-17 % ) and Upwork ( 19 % ).

While consumer EdTech companies like Chegg (-62 % ) and CourseHero (-68 % ) are in free fall, search engines like Yahoo (-12 % ) and Bing (-14 % ) continue a multi-quarter decline in visits.

The evidence suggests that early-stage AI disruption is beginning to lessen the usefulness of some legacy platforms. Additionally, it provides a hint to businesses that use or create these services: the time may be right to lessen their reliance on them, from a revenue-generating, marketing, or overall business perspective.

With VB Daily, you get daily insights on business use cases.

VB Daily has you covered if you want to impress your boss. In order to provide you with the most insightful information on what businesses are using generative AI, from regulatory changes to real-world deployments, we give you the inside information on what companies are doing with it.

Read our privacy statement.

Thank you for your subscription. View more VB newsletters here.

An error took place.