Coding agents see a 75% surge: SamelyWeb’s AI Usage Report highlights the departments that generate AI BOOM wins and loses in 2025

As generative AI continues to redefine digital workflows across industries, SameWeb’s “AI Global Report: Trends in Generating AI Globally” (End on May 9, 2025) A comprehensive snapshot of the transfer user participation model is provided. The data-driven report highlights the significant growth in coding proxy, the destructive impact on Edtech, and the unexpected recession of legal AI platforms. Here are five findings that stand out from the report’s multi-departmental analysis.

1. AI-driven coding tools witness continued momentum

In the highest performing category DEVOPS and code completion Recording tools 75% year-on-year (young) Transportation increased. This growth reflects the increase in adoption of developers’ AI assistants that support code generation, error detection, and workflow automation.

Two platforms are prominent: cuteThere is an amazing one Year-on-year growth of 207%and cursor,register Increased by 62%. These tools gain traction due to their seamless integration with IDE and DEVOPS pipelines, reducing developer cognitive overhead and enhancing speed in iterative software engineering environments.

2. Universal LLM destroys traditional Edtech models

Chat-based LLM, e.g. Openai’s Chatgpt,,,,, DeepSeekand Groke Has become a central tool for self-guided learning and on-demand coaching. DeepSeek, in particular, experienced exponential traffic spikes 17,000% year-on-year– Before adjustments are made in May.

This surge coincides with the significant decline in traditional education platforms. this Edtech Category experience 24% traffic droplegacy player chegg and Course Hero Posting drops sharply -62% and -68%respectively. Data show that LLM effectively places static repositories with dialogue, real-time educational support (especially for STEM and writing tasks).

3. Legal AI tools enter economic downturn when using fatigue

In contrast to the buoyancy of the encoding tool, Legal AI The segment is facing a significant contraction. 73% drop In traffic. This decline may reflect saturation in niche markets where the value proposition of generative AI (contract summary, legal drafting, compliance automation) is not yet fully matured into a strong enterprise-level deployment.

The data show that despite strong early interest in legal AI, retention and continuous use remain challenging. Legal practitioners may postpone wider adoption until the tools are better aligned with real-world legal reasoning, jurisdictional nuances and auditability requirements.

4. Video generation tool delivers mixed signals

this Video generation Department only displays -5% change Overall, however, this average masking is significant platform-specific differences. kling.ai and track See traffic drop 5% and 15%although Hagen Recorded Increased by 25%– Very attributed to its focus on addressing specific commercial use cases, such as synthetic avatars for commercial communications.

This difference highlights a broader trend: Video integrated platforms that cannot meet clear market demands or lack intuitive UI/UX are working to maintain user interest. In contrast, those who are consistent with corporate storytelling or content automation will see more durable engagement.

5. Free platform feels the pressure of AI automation

The report also highlights 17% year-on-year In traffic Digital Freelancer platform. Fiverr and Upwork Specially affected, decreased 15% and 19%respectively. The underlying drivers appear to be the growing ability of generative AI to automate traditional free-driven tasks (copy, basic design, SEO analysis and transcription) to shift demand from manual labor.

The free economy may be entering a transitional phase of successful dependence on human collaboration. Freelancers who adapt by providing AI-enhanced services or specializing in areas that require nuanced judgment may find new opportunities as others contract.

in conclusion

Similar Networks’ Global Report on AI: Trends in Generative AI Globally explores the forks of Generative AI adoption: Platforms (especially in development and operations) that address specific areas of challenges through measurable productivity gains (especially in development and operations). Instead, tools that lack differentiation or have not yet proven to be real reliability are witnessing the loss.

As AI continues to be more deeply integrated into professional toolchains, user engagement is increasingly driven by goals and ROI. This is not only a technological shift, but also a redefinition of the digital productivity landscape across sectors.

Download the report. All credits for this study are to the researchers on the project. Also, please stay tuned for us twitter And don’t forget to join us 90K+ ml reddit.