Enterprise AI Adoption Accelerates, Widening Leader-Laggard Gap
TL;DR
- Enterprise AI adoption is accelerating rapidly, with coding emerging as the primary killer use case, driving significant ROI and enabling new capabilities beyond mere efficiency gains.
- The gap between AI leaders and laggards is widening, as early adopters compound their advantages through systematic investment in AI infrastructure and operating models.
- Reasoning token consumption per organization has increased dramatically, indicating a shift towards more complex and sophisticated AI-driven workflows beyond basic efficiency.
- Companies are increasingly preferring to purchase AI solutions rather than build them internally, reflecting the maturity of AI application layer startups and vendor offerings.
- Open-source LLMs are declining in enterprise adoption, with companies showing wariness towards Chinese models and a preference for closed-source, performant alternatives.
- While agents are a future focus, current enterprise AI spend is heavily skewed towards copilots, indicating that simpler AI architectures still dominate production deployments.
Deep Dive
Enterprise AI adoption is accelerating dramatically, with coding emerging as the primary killer use case and reasoning models driving deeper workflow integration. This rapid expansion is widening the gap between early adopters who are compounding their advantages and laggards, signaling a potential "boom" rather than a "bubble" as AI fundamentally reshapes enterprise operations beyond mere efficiency gains.
Two recent reports, OpenAI's "State of Enterprise AI" and Menlo Ventures' "State of Generative AI in the Enterprise," highlight key trends. OpenAI's data, derived from enterprise customer usage and a survey of 9,000 workers, shows a 900% year-over-year increase in ChatGPT Enterprise seats and an 800% rise in weekly enterprise messages. Crucially, this growth is not just in frequency but also in depth, with custom GPTs and projects--used for repeatable, multi-step tasks--growing 19x, indicating deeper integration into core workflows. Industries like technology, healthcare, and manufacturing are leading this charge, experiencing 11x, 8x, and 7x growth, respectively. The adoption of reasoning tokens has exploded by 320x, enabling more complex work. Significantly, 75% of surveyed workers report improved speed or quality of output, with savings of 40-60 minutes daily, and 87% of IT workers report faster resolutions. Beyond efficiency, AI is enabling new capabilities; 75% of workers can now complete tasks previously impossible, particularly in areas like coding, spreadsheet analysis, and technical tool development. Coding-related messages outside traditional engineering functions have grown by an average of 36% in six months, suggesting a democratization of development capabilities. The reports also reveal a widening disparity between "frontier" individuals and firms, who consume significantly more AI resources and achieve greater time savings, and the median user, indicating that early and intensive adopters are gaining a disproportionate advantage.
Menlo Ventures' report, surveying 495 enterprise AI decision-makers, corroborates these findings, labeling generative AI as the fastest-scaling software category in history, with $37 billion in spend this year. Coding is confirmed as the killer app, accounting for 55% of departmental AI spend, totaling $4 billion. AI app builders have seen a 10x increase in enterprise spend, and code agents a 36.7x surge. The report identifies Anthropic as the current enterprise AI leader, capturing 40% of enterprise LLM spend, followed by Google, while OpenAI's share has decreased. Interestingly, startups are capturing significant revenue in the application layer, earning roughly two dollars for every dollar earned by incumbents, a shift from previous technology waves where incumbents typically held an advantage. The "buy versus build" paradigm also shows a strong preference for purchasing AI solutions, with 76% of use cases being bought rather than built internally, suggesting a maturation of the vendor landscape. While excitement around agents is high, copilot usage represents ten times more enterprise spend, indicating that simpler AI augmentation tools are currently more prevalent than fully autonomous systems, with only 16% of enterprise deployments qualifying as true agentic systems.
The core implication is that AI is no longer just an efficiency tool but a fundamental driver of new capabilities and competitive differentiation. Enterprises that systematically invest in AI infrastructure and integrate it deeply into their operations are achieving compounding advantages, leading to faster growth and a widening chasm between leaders and laggards. The rise of coding as a primary use case democratizes development and amplifies productivity across various departments, while the increasing preference for purchasing AI solutions points to a developing ecosystem where specialized vendors can thrive.
Action Items
- Audit enterprise AI adoption: Analyze 3-5 key workflows for AI integration depth and identify leaders versus laggards.
- Create custom GPTs: Develop 3-5 specialized GPTs for common repeatable tasks, focusing on workflow integration.
- Measure AI-driven coding impact: Track coding-related message growth in non-technical departments by 30% over 6 months.
- Evaluate AI vendor partnerships: Compare revenue capture of AI startups versus incumbents in the application layer for 3-5 categories.
- Assess agentic system maturity: For 3-5 enterprise deployments, evaluate the complexity of AI architectures beyond fixed sequences.
More from The AI Daily Brief: Artificial Intelligence News and Analysis
Task Imagination Unlocks Fable 5's Autonomous Delegation
The bottleneck with Fable 5 isn't model capability; it's your ability to imagine tasks that take hours or days. The real advantage goes to those who develop "task imagination," delegating responsibilities rather than simple commands.
View Episode Notes →
AI Research Automation Is the New Battleground for Technological Dominance
AI isn't just changing products--it's automating its own evolution. The real race isn't for users, but for control over the pace of progress, where AI research conducted by AI creates a compounding advantage few can match.
View Episode Notes →
Orchestrating AI Loops Creates Compounding Advantage
AI's real power isn't in prompts--it's in loops. Users who treat AI as a reasoning partner, not a chatbot, gain compounding advantages through automation, while others hit diminishing returns. The gap is widening.
View Episode Notes →Key Quotes
"Altman told the journal that two models are planned garlic or gpt 5 2 and another model in january gpt 5 2 will deliver a boost in capabilities for ai coders and enterprise customers as well as hopefully generally build some momentum the january model is intended to have better images and personality and the code red will supposedly end upon its release take all that together and it means that for eight weeks all work on store development the focus on agi all of that stuff has been put aside in favor of improving the chat gpt experience"
This quote highlights OpenAI's strategic shift, prioritizing immediate user experience improvements for ChatGPT over long-term Artificial General Intelligence (AGI) development. Sam Altman's framing suggests an existential need to deliver current value to ensure the company's survival. The planned models, GPT-5.2 and a January release, aim to enhance capabilities for coders and enterprise customers, with the latter focusing on improved imagery and personality.
"The neutral organization was necessary to ensure that agents and systems work together without competing standards he commented we need multiple protocols to negotiate communicate and work together to deliver value for people and that sort of openness and communication is why it's not ever going to be one provider one host one company"
Nick Cooper, an OpenAI engineer, explains the rationale behind establishing a neutral organization for agentic AI. He emphasizes the need for interoperability, stating that multiple protocols must negotiate and communicate to deliver value. Cooper's comment suggests that a single provider or host controlling all agents is unlikely due to the necessity of open standards for collaboration.
"The gap between leaders and laggards is increasing because what constitutes that gap is making the leaders grow faster than the laggards there's some evidence of this on both the individual and the organization level in this report basically chatgpt found when it came to individuals the people who consumed the most intelligence as measured by credits used reported higher time savings the group that saved over 10 hours a week used eight times more credits than the group who reported saving zero hours a week"
This quote from the OpenAI report illustrates the widening disparity between AI adoption leaders and laggards. The author points out that individuals and organizations using AI more intensively report greater benefits, such as increased time savings. Specifically, those saving over 10 hours weekly used significantly more AI credits than those reporting no savings, indicating a direct correlation between adoption depth and realized value.
"Menlo calls it the fastest scaling software category in history which at 37 billion in spend this year captures 6 of the 300 billion global sas market just three years after chatgpt was released"
This statement from the Menlo Ventures report characterizes generative AI as the fastest-scaling software category ever. The author notes that in just three years since ChatGPT's release, generative AI spending has reached $37 billion, representing 6% of the global Software as a Service (SaaS) market. This rapid growth underscores the significant and swift impact AI is having on the enterprise software landscape.
"Menlo argues that at this point anthropic is fairly definitively the enterprise ai leader they estimate that anthropic ai now earns 40 of enterprise llm spend which was up from 24 in 2024 and 12 in 2023 google also saw a big jump going from 7 in 23 to 21 in 25 and all of this happened at the expense a little bit of meta which went from 16 in 23 to 8 in 25 and a lot at openai which went from 50 in 23 down to 27 in 25"
According to Menlo Ventures, Anthropic has emerged as the leading enterprise AI provider, capturing 40% of enterprise Large Language Model (LLM) spending in 2025. This represents a substantial increase from previous years. The author also notes significant gains for Google and a decline for Meta and OpenAI in enterprise LLM market share, indicating a shifting competitive dynamic.
"Menlo saw a flip flopping in 2024 64 of enterprises said they preferred buying software versus just 20 building but that shifted dramatically in 2024 when they found 47 of solutions being developed in house with 53 being sourced from vendors now i argued at the time that what that reflected was one a growing confidence of enterprises in engaging with ai but two the immaturity of vertical and departmental and functional level startups that i thought was going to change pretty quickly it just seemed to me that even with increased confidence and even with the decreased cost of building software once highly focused application layer startups came online for key use cases i thought that was going to boomerang back to about that 80 20 split sure enough in this study they found 76 of ai use cases being purchased rather than built internally"
This quote from the Menlo report details a significant shift in enterprise AI development strategy. While in 2024, enterprises leaned towards building solutions in-house, the 2025 study shows a strong return to purchasing AI use cases from vendors, with 76% of solutions being bought rather than built. The author suggests this boomerang effect is due to the increasing maturity of specialized AI startups offering focused solutions.
Resources
External Resources
Books
- "The State of Enterprise AI" by OpenAI - Mentioned as a report analyzing enterprise AI adoption, usage data, and survey results.
- "Third Annual State of Generative AI in the Enterprise" by Menlo Ventures - Mentioned as a report surveying enterprise AI decision-makers on spending, adoption, and market trends.
Articles & Papers
- "Code Red Plans" (Wall Street Journal) - Mentioned as a write-up detailing OpenAI's strategic plans for model releases.
People
- Amar - Product and design lead at Google DeepMind, mentioned in relation to vibe coding with Gemini 3.
- Nick Cooper - OpenAI engineer, commented on the necessity of neutral organizations for agent negotiation and communication.
- Mike Krieger - Anthropic's Chief Product Officer, discussed the stewardship of the Model Context Protocol (MCP) by the Agentic AI Foundation.
- Pete Hegseth - Secretary of War, commented on the US military's unveiling of the GenAI.mil platform.
- Emil Michael - Under Secretary for Research and Engineering, commented on the US military's rapid deployment of AI capabilities.
Organizations & Institutions
- OpenAI - Mentioned for releasing a report on enterprise AI, donating the Agent MD instruction format to the Agentic AI Foundation, and launching AI Foundations certification courses.
- Menlo Ventures - Mentioned for releasing their third annual report on generative AI in the enterprise.
- Google - Mentioned for its Gemini for Government platform being hosted on the US military's GenAI.mil platform.
- Anthropic - Mentioned for donating the Model Context Protocol (MCP) to the Agentic AI Foundation and for leading enterprise AI according to the Menlo report.
- Block - Mentioned as a co-founder of the Agentic AI Foundation, donating their Goose agent framework.
- Linux Foundation - Mentioned as the host for the Agentic AI Foundation, ensuring its independence.
- Walmart - Mentioned as a partner for OpenAI's AI Foundations certification courses.
- John Deere - Mentioned as a partner for OpenAI's AI Foundations certification courses.
- Lowe's - Mentioned as a partner for OpenAI's AI Foundations certification courses.
- BCG - Mentioned as a partner for OpenAI's AI Foundations certification courses.
- Accenture - Mentioned as a partner for OpenAI's AI Foundations certification courses.
- Arizona State University - Mentioned as a university piloting OpenAI's AI Foundations certification courses.
- California State University - Mentioned as a university piloting OpenAI's AI Foundations certification courses.
- US Department of War - Mentioned for unveiling the GenAI.mil AI platform for the military.
- Superintelligent - Mentioned as the company behind the "Plateau Breaker" assessment and the "Agent Readiness Audit."
- Rovo - Mentioned as a provider of AI-powered search, chat, and agents.
- AssemblyAI - Mentioned as a provider for building Voice AI apps.
- LandfallIP - Mentioned as a provider for navigating the patent process with AI.
- Blitzy.com - Mentioned as a platform for building enterprise software.
- Robots & Pencils - Mentioned as a provider of cloud-native AI solutions and applied experimentation.
- Atlassian - Mentioned as the platform provider for Robo, an AI-powered teammate.
- KPMG - Mentioned as a sponsor of the podcast, with a podcast titled "You Can with AI."
Tools & Software
- Gemini 3 - Mentioned as a tool in Google AI Studio for building apps without coding background.
- GPT-5.2 - Mentioned as a rumored upcoming model from OpenAI.
- GPT Image Model (code named Chestnut and Hazelnut) - Mentioned as rumored new image models being tested by OpenAI.
- Agent MD - Mentioned as an instruction format donated by OpenAI to the Agentic AI Foundation.
- Goose Agent Framework - Mentioned as a framework donated by Block to the Agentic AI Foundation.
- Model Context Protocol (MCP) - Mentioned as a protocol donated by Anthropic to the Agentic AI Foundation.
- ChatGPT Enterprise - Mentioned as a product with significant year-over-year growth in enterprise seats and messages.
- Custom GPTs and Projects - Mentioned as interfaces for interacting with ChatGPT that are growing faster than overall usage.
- Gemini for Government - Mentioned as the first AI service to be hosted on the US military's GenAI.mil platform.
- Robo - Mentioned as an AI-powered teammate for search, chat, and agents, built on Atlassian's platform.
- Plateau Breaker - Mentioned as a new assessment from Superintelligent to break through AI plateaus.
- Agent Readiness Audit - Mentioned as an offering from Superintelligent.
- Blitz AI - Mentioned as an enterprise autonomous software development platform.
Courses & Educational Resources
- AI Foundations - Mentioned as the first set of certification courses launched by OpenAI, in collaboration with Coursera.
- ChatGPT Foundations - Mentioned as a course for teachers on using ChatGPT.
Websites & Online Resources
- ai.studio/build - Mentioned as the URL to visit to create an app with Gemini 3 in Google AI Studio.
- rovo.com - Mentioned as the URL for Rovo, an AI-powered teammate.
- assemblyai.com/brief - Mentioned as the URL for AssemblyAI.
- landfallip.com - Mentioned as the URL for LandfallIP.
- blitzy.com - Mentioned as the URL for Blitzy.com.
- robotsandpencils.com - Mentioned as the URL for Robots & Pencils.
- besuper.ai - Mentioned as the URL for Superintelligent, to request an agent readiness score.
- patreon.com/aidailybrief - Mentioned as a way to get an ad-free version of the show.
- pod.link/1680633614 - Mentioned as a link to subscribe to the podcast version of The AI Daily Brief.
Podcasts & Audio
- The AI Daily Brief: Artificial Intelligence News and Analysis - Mentioned as the podcast hosting the episode.
- KPMG 'You Can with AI' podcast - Mentioned as a new podcast from KPMG.
Other Resources
- Coding - Mentioned as the first true killer use case for enterprise AI adoption.
- Reasoning Models - Mentioned as driving deeper workflow integration in enterprise AI.
- Agentic AI Foundation - Mentioned as a newly established foundation by OpenAI, Anthropic, and Block to foster innovation in agentic AI.
- Vibe Coding - Mentioned as a method to build apps with Gemini 3 in Google AI Studio by describing the app.
- Agentic AI Systems - Mentioned as a focus for development within the Agentic AI Foundation.
- AI ROI Benchmarking Study - Mentioned as a study with over 5,000 use cases being analyzed.
- Boom-versus-bubble debate - Mentioned as a dominant conversation in market analysis regarding AI.
- SAS Market - Mentioned as a market where generative AI spend is capturing a significant percentage.
- Horizontal AI - Mentioned as a category of AI spend that is growing.
- Departmental AI - Mentioned as a category of AI spend that is growing.
- Vertical AI - Mentioned as a category of AI spend that is growing.
- Application Layer AI - Mentioned as a significant area of spend and startup success.
- Buy-build paradigm - Mentioned as being blurry in the context of AI, with enterprises purchasing more AI use cases than building them.
- Open Source LLMs - Mentioned as seeing a decline in enterprise adoption.
- Llama Models - Mentioned as potentially stagnating, contributing to the decline of open source LLMs in enterprise.
- Copilots - Mentioned as representing significantly more enterprise spend than agents.
- Modern AI Stack - Mentioned as still being in development.
- Agentic Systems - Mentioned as a category of AI deployment, with a low percentage of enterprise deployments qualifying as true agentic systems.
- Fixed Sequence Workflows - Mentioned as a type of AI deployment, contrasted with multi-agent systems.
- Multi-Agents - Mentioned as a type of AI deployment.