Research analyzing four million conversations with Claude AI has uncovered a striking concentration in how artificial intelligence gets deployed across work tasks. Just 5% of occupational tasks account for 59% of all AI interactions, according to a paper published October 29, 2025, authored by researchers from Adobe and Netaji Subhas University of Technology.

The findings challenge widespread assumptions about which marketing activities benefit most from AI augmentation. Tasks demanding high creativity, cognitive complexity, and idea generation attract the heaviest AI engagement. Routine work shows minimal adoption despite years of predictions that AI would primarily automate repetitive activities.

The research team—Peeyush Agarwal, Harsh Agarwal from Adobe, and Akshat Rana—analyzed data from Anthropic's Economic Index, which mapped anonymized user interactions with the Claude assistant to standardized occupational tasks from the U.S. Department of Labor's O*NET database. The dataset covered December 2024 through January 2025, examining approximately one million conversations on Claude.ai's Free and Pro plans.

Marketing professionals represent a significant portion of this usage pattern. IAB Europe revealed on September 18, 2025, that 85% of companies already deploy AI-based tools for marketing purposes, with content generation and targeting leading adoption at 61% and 64% respectively. The concentrated usage patterns revealed in this new research suggest substantial variation in which specific marketing tasks benefit from AI assistance.

The characteristics that drive adoption

The researchers developed a comprehensive framework evaluating each task across seven dimensions: Routine, Cognitive, Social Intelligence, Creativity, Domain Knowledge, Complexity, and Decision Making. Each dimension broke down into five specific parameters, creating 35 measurable characteristics scored using large language models.

Tasks demanding idea generation showed the strongest correlation with AI usage, with a Spearman correlation coefficient of 0.173. Information processing followed at 0.157, while originality requirements correlated at 0.151. Conversely, tasks characterized by predictable outcomes showed negative correlation at -0.135, while frequency of repetition correlated negatively at -0.131.

The pattern indicates AI adoption concentrates in divergent and information-intensive phases of work rather than convergent implementation. Within the Creativity dimension, idea generation correlated much more strongly with AI usage than innovation requirements. Tasks requiring artistic or aesthetic components showed lower correlation than pure conceptual work.

High-usage tasks—defined as the top 10th percentile—exhibited a distinct signature compared to low-usage tasks in the bottom 10th percentile. High-usage tasks scored substantially higher on Cognitive dimensions at 8.8 versus 6.8, Complexity at 8.7 versus 6.5, Creativity at 7.1 versus 2.8, and Decision Making at 8.3 versus 6.6. They scored significantly lower on Routine at 2.9 versus 6.2.

Notably, Social Intelligence showed minimal difference between groups at 6.1 versus 5.7. This finding suggests that interpersonal skill requirements neither attract nor repel AI adoption in current usage patterns. For marketing teams, this indicates that customer-facing activities requiring empathy and relationship management remain primarily human domains despite AI's expanding capabilities.

Three distinct task archetypes emerge

The researchers employed Principal Component Analysis to identify underlying dimensions, finding the first two components explained 82.3% of variance across all task characteristics. K-Means clustering on these components identified three distinct task archetypes, validated through Multivariate Analysis of Variance showing the clusters were highly statistically distinct.

Dynamic Problem Solving tasks, comprising 2,100 tasks or the largest group, attracted the highest mean AI usage at 3.31%. These tasks scored lowest on routineness at 3.36 and highest on cognitive demands at 8.53, creativity at 6.43, complexity at 8.43, and decision making at 8.25. The archetype represents work requiring substantial intellectual flexibility and original thinking.

Procedural & Analytical Work, consisting of 1,017 tasks, showed moderate AI usage at 2.45%. These tasks demonstrated moderate routineness at 5.8, low social intelligence at 4.51, and low creativity at 3.39. The profile suggests structured analytical activities that follow established frameworks but require cognitive effort.

Standardized Operational Tasks, the smallest group with 397 tasks, exhibited the lowest AI usage at 0.014%. These tasks scored highest on routineness at 7.08 and lowest across all other dimensions, including cognitive demands at 4.6, social intelligence at 3.1, and creativity at 1.53.

Within each archetype, usage distribution proved extremely right-skewed. The median usage representing the typical task fell substantially lower than the mean. The key distinction appeared in the nature of the upper tail: Dynamic Problem Solving tasks contained both more high-usage outliers and outliers of greater magnitude than other archetypes.

This pattern indicates that higher average usage for complex tasks reflects concentrated, "spiky" application of AI to select activities rather than broad increases in AI attention across all demanding work. Marketing campaign conceptualization and strategic planning align with Dynamic Problem Solving characteristics. Standard reporting and repetitive media buying operations show characteristics associated with Standardized Operational Tasks.

Cognitive offloading as the central phenomenon

The research identifies cognitive offloading as the psychological dimension driving current AI adoption. Individuals leverage AI to overcome initial high-friction stages of knowledge work, particularly brainstorming, outlining, and synthesizing information. The pattern appears across all task categories but manifests most strongly in complex cognitive work.

Domain knowledge requirements showed moderate positive correlation with AI usage at 0.084, suggesting specialized expertise neither strongly attracts nor repels AI adoption. This finding carries implications for marketing specializations. Practitioners with deep expertise in programmatic advertising or conversion rate optimization can apply AI tools effectively, but domain knowledge itself doesn't predict whether they will adopt these tools.

The findings align with observations documented on October 26, 2025, when technical practitioners questioned whether AI agents get overused for tasks where simpler code would perform better. Multiple participants argued that AI agents prove valuable for problems requiring pattern recognition and unstructured language interpretation, but traditional programming remains appropriate for core infrastructure requiring deterministic logic.

Marketing measurement challenges may explain some adoption patterns. Research released December 2, 2025, by Funnel and Ravn Research found that 86% of in-house marketers struggle to determine the impact of each marketing channel on overall performance despite unprecedented access to analytics tools. AI tools that synthesize information and generate insights address these information processing challenges that characterize high-usage tasks.

Enterprise adoption faces scaling challenges

McKinsey released research on November 9, 2025, revealing that while 88% of respondents report regular AI use in at least one business function, only approximately one-third report their companies have begun scaling AI programs across the enterprise. The concentrated usage patterns revealed in this research provide potential explanation: most tasks may not yet exhibit characteristics that attract heavy AI adoption.

Marketing organizations face decisions about which specific activities warrant AI investment. The research framework provides evidence-based guidance. Tasks combining high creativity requirements with substantial cognitive complexity but low routineness represent the highest-probability AI adoption candidates. Standard operational activities may require different automation approaches than conversational AI interfaces.

Anthropic released its Economic Index on February 10, 2025, revealing that approximately 36% of occupations use AI for at least a quarter of their associated tasks, while only 4% of occupations utilize AI across three-quarters of their tasks. Computer-related tasks saw the largest AI usage at 37.2% of all queries, followed by writing tasks in educational and communication contexts at 10.3%.

The research challenges assumptions about AI replacing routine work. While earlier automation technologies primarily targeted repetitive tasks, current generative AI adoption shows opposite patterns. Tasks with high routineness and predictable outcomes correlate negatively with usage, while cognitively demanding creative work correlates positively.

Implications for marketing operations

Marketing teams serving younger demographics can leverage advanced AI tools for content creation and campaign optimization. German research published August 5, 2025, showed younger audiences demonstrate high comfort levels with AI-powered experiences and may expect sophisticated personalization and automation. Older demographics show limited AI engagement and may prefer traditional marketing approaches.

The concentrated usage patterns suggest marketing organizations should systematically analyze jobs requiring high cognitive complexity and creativity, then develop AI-powered processes around these functions. These represent the tasks that data shows people readily offload and constitute clear market signals for where AI adoption will accelerate.

Performance benefits vary significantly across market segments. IAB Europe data showed ad tech firms report the strongest results, with 60% citing KPI improvements from AI adoption. Agencies follow at 48%, while publishers remain more cautious with fewer than one-third reporting CPM increases. This disparity suggests different value propositions across the advertising value chain.

Technical applications span multiple advertising functions beyond targeting and content generation. Companies utilize AI for programmatic optimization, dynamic creative optimization, audience segmentation, reporting and analysis, personalized experiences, media planning, measurement, cross-channel optimization, and brand safety checks.

The research identifies significant skills gaps hindering broader AI adoption. Lack of internal expertise and training represents the primary barrier, cited by 45% of respondents in the IAB Europe study. Integration difficulties with existing systems and regulatory uncertainty follow as secondary obstacles, each mentioned by 33% of companies.

Governance challenges persist despite widespread adoption

The data reveals concerning governance gaps despite widespread adoption. While 68% of companies maintain general internal AI guidelines, only 43% have developed marketing-specific frameworks according to IAB Europe research. Nearly one-fifth of organizations lack formal AI governance entirely, with most responsibility falling to dedicated leads rather than distributed teams or ethics committees.

User data policies for AI training remain divided across the ecosystem. The survey found 43% of companies prohibit using user data for AI model training, while 32% allow internal use only. Seven percent permit third-party training partnerships, and 17% enable both internal and external data usage. These divergent approaches reflect ongoing uncertainty about data governance best practices.

Third-party auditing of AI tools remains limited. Only 16% of companies conduct external assessments of their AI systems, while 33% rely solely on internal evaluation. Training initiatives attempt to address knowledge deficits. Sixty percent of companies provide AI education to marketing personnel, while two-thirds express interest in industry association guidelines for AI technology usage.

McKinsey data published July 27, 2025, indicates $1.1 billion in equity investment flowed into agentic AI in 2024, with job postings related to this technology increasing 985 percent from 2023 to 2024. The report notes that while interest levels remain relatively low compared to established AI technologies, growth rates exceed all other technology trends.

Future research directions

The researchers acknowledged several limitations. The usage data derives from a single AI model family, Claude, whose user base may not represent the entire workforce. The LLM-based scoring method for task characteristics, while systematically applied, serves as a proxy for human judgment and may carry inherent biases. The analysis provides a cross-sectional snapshot rather than capturing dynamic evolution of AI use as technology advances and adoption patterns change.

Future research should pursue validation through comparable usage data from other major AI labs and longitudinal studies tracking the evolution of adoption patterns over time. Investigation of the psychological and organizational mechanisms driving the observed preference for delegating complex cognitive work would deepen understanding. Comparison of actual AI adoption patterns with measured AI capabilities across task characteristics could identify potential misalignment between usage and optimal deployment.

The paper introduces three key contributions to understanding AI's impact on work. First, systematic evidence linking real-world AI usage patterns to intrinsic task characteristics. Second, a comprehensive multi-dimensional framework for characterizing occupational tasks that captures the multifaceted nature of modern AI capabilities. Third, distinct task archetypes revealing deeper patterns in AI adoption than individual characteristics alone.

Future implications span multiple domains. Labor markets may experience fundamental transformation as knowledge work shifts from information processing toward task delegation, decision making, and quality evaluation. Educational priorities require reorientation toward critical thinking, task delegation capabilities, and analytical skills to distinguish between valuable AI contributions and potential errors.

The evidence suggests that individuals most likely to benefit are those who can effectively navigate the partnership between human judgment and AI capability, applying their expertise to direct, refine, and contextualize AI's output rather than attempting to replicate its cognitive processing power. Social intelligence maintains a distinct position in this landscape, remaining statistically decoupled from AI adoption patterns and suggesting that interpersonal capabilities continue to represent an area of human comparative advantage.

Timeline

Summary

Who: Researchers Peeyush Agarwal from Netaji Subhas University of Technology, Harsh Agarwal from Adobe Inc., and Akshat Rana from Netaji Subhas University of Technology conducted the analysis, utilizing Anthropic's Economic Index dataset of Claude AI interactions.

What: The research paper reveals extreme concentration in AI usage patterns, with just 5% of occupational tasks accounting for 59% of all interactions with Claude AI, demonstrating that tasks requiring high creativity, complexity, and cognitive demand but low routineness attract the most engagement while highly routine work remains largely untouched by generative tools.

When: The paper was published on arXiv on October 29, 2025 (version 2), analyzing conversation data from December 2024 and January 2025, following initial submission on October 26, 2025.

Where: The research analyzed data from global Claude.ai users across multiple occupational categories using the standardized O*NET task taxonomy from the U.S. Department of Labor, with findings applicable to marketing professionals and knowledge workers worldwide navigating AI integration in their workflows.

Why: Understanding which intrinsic task characteristics drive AI adoption enables marketing organizations to make evidence-based decisions about technology investment, helps educational institutions develop relevant curricula for AI-augmented work environments, and provides policymakers with frameworks for facilitating labor market transitions as cognitive work fundamentally transforms from information processing toward task delegation and quality evaluation.

Share this article
The link has been copied!