Another week on the road meeting with a couple dozen IT and AI leaders from large enterprises across banking, media, retail, healthcare, consulting, tech, and sports, to discuss agents in the enterprise. Some quick takeaways: * Clear that we’re moving from chat era of AI to agents that use tools, process data, and start to execute real work in the enterprise. Complementing this, enterprises are often evolving from “let a thousand flowers bloom” approach to adoption to targeted automation efforts applied to specific areas of work and workflow. * Change management still will remain one of the biggest topics for enterprises. Most workflows aren’t setup to just drop agents directly in, and enterprises will need a ton of help to drive these efforts (both internally and from partners). One company has a head of AI in every business unit that roles up to a central team, just to keep all the functions coordinated. * Tokenmaxxing! Most companies operate with very strict OpEx budgets get locked in for the year ahead, so they’re going through very real trade-off discussions right now on how to budget for tokens. One company recently had an idea for a “shark tank” style way of pitching for compute budget. Others are trying to figure out how to ration compute to the best use-cases internally through some hierarchy of needs (my words not theirs). * Fixing fragmented and legacy systems remain a huge priority right now. Most enterprises are dealing with decades of either on-prem systems or systems they moved to the cloud but that still haven’t been modernized in any meaningful way. This means agents can’t easily tap into these data sources in a unified way yet, so companies are focused on how they modernize these. * Most companies are *not* talking about replacing jobs due to agents. The major use-cases for agents are things that the company wasn’t able to do before or couldn’t prioritize. Software upgrades, automating back office processes that were constraining other workflows, processing large amounts of documents to get new business or client insights, and so on. More emphasis on ways to make money vs. cut costs. * Headless software dominated my conversations. Enterprises need to be able to ensure all of their software works across any set of agents they choose. They will kick out vendors that don’t make this technically or economically easy. * Clear sense that it can be hard to standardize on anything right now given how fast things are moving. Blessing and a curse of the innovation curve right now - no one wants to get stuck in a paradigm that locks them into the wrong architecture. One other result of this is that companies realize they’re in a multi-agent world, which means that interoperability becomes paramount across systems. * Unanimous sense that everyone is working more than ever before. AI is not causing anyone to do less work right now, and similar to Silicon Valley people feel their teams are the busiest they’ve ever been. One final meta observation not called out explicitly. It seems that despite Silicon Valley’s sense that AI has made hard things easy, the most powerful ways to use agents is more “technical” than prior eras of software. Skills, MCP, CLIs, etc. may be simple concepts for tech, but in the real world these are all esoteric concepts that will require technical people to help bring to life in the enterprise. This both means diffusion will take real work and time, but also everyone’s estimation of engineering jobs is totally off. Engineers may not be “writing” software, but they will certainly be the ones to setup and operate the systems that actually automate most work in the enterprise.
One-page infographic (2025) titled 'The Rise of Agentic AI' that diagrams the evolution from workflow engines/RPA to orchestrated autonomous agents, includes adoption stats and a roadmap of capabilities (data management, connectors, agent engineering, orchestration). It directly supports the takeaways by highlighting the shift to agentic automation, the need for modernized data/platform foundations and multi-agent orchestration, and the enterprise concerns around integration, governance, and operational readiness.
Source: Informatica
Research Brief
What our analysis found
The enterprise AI landscape is undergoing a decisive architectural shift — from simple chatbot interfaces to autonomous agents capable of using tools, processing data, and executing complex workflows. Gartner predicted in March 2024 that by 2028, one-third of all interactions with generative AI services will rely on action models and autonomous agents for task completion, a forecast now being validated by major vendor moves from Microsoft, Snowflake, and others racing to productize agent runtimes and orchestration layers. Conversations with IT and AI leaders across banking, media, retail, healthcare, and other sectors confirm that enterprises are pivoting from experimental "let a thousand flowers bloom" AI adoption toward targeted, workflow-specific automation projects — but deep organizational and technical barriers remain.
Among the most pressing challenges is the sheer cost of running agentic AI at scale. Enterprises operating under strict annual OpEx budgets are now engaged in intense trade-off discussions about how to allocate compute and token spend, with some companies reportedly experimenting with internal "shark tank"-style pitch competitions to ration AI budgets to the highest-value use cases. Meanwhile, legacy and fragmented IT systems — often decades old and never meaningfully modernized after cloud migration — remain a top blocker, preventing agents from accessing enterprise data in a unified way. CIO surveys from Gartner, Forrester, and IDC consistently rank system modernization and composable API architectures among the top digital priorities for 2025 and 2026.
Perhaps most notable is the emerging consensus around interoperability and the rising importance of engineering talent. The Linux Foundation launched the Agentic AI Foundation in late 2025, taking stewardship of Anthropic's Model Context Protocol (MCP) and other early interoperability standards, while Google introduced its Agent-to-Agent (A2A) protocol in April 2025. These moves signal that enterprises expect to operate in a multi-agent, multi-vendor world — and that the technical complexity of deploying agents (skills, MCP servers, CLI tooling) will sustain or even increase demand for engineering roles, even as the nature of software development itself transforms. The World Economic Forum's 2025 Future of Jobs Report found that roughly 41% of employers plan workforce downsizing in roles susceptible to AI automation, but many of those same organizations are simultaneously investing heavily in upskilling — suggesting the real story is workforce transformation, not wholesale replacement.
Fact Check
Evidence from both sides
Supporting Evidence
Gartner validates the shift from chat to agentic AI
In March 2024, Gartner forecast that by 2028, one-third of interactions with generative AI services will use action models and autonomous agents for task completion, directly supporting the tweet's claim that the industry is moving beyond simple chat interfaces toward tool-using, work-executing agents.
Major vendor product launches confirm the agentic trend
Microsoft's announcements at Ignite 2024 and subsequent 2025 product releases — including Copilot Studio, Azure Foundry agent runtimes, and model routing for cost and latency optimization — demonstrate that the largest enterprise software vendors are building production-grade agent infrastructure, as reported by AP News and Microsoft's own engineering blogs.
Change management is consistently cited as a top adoption barrier
Analyst reports from Gartner, Deloitte, and McKinsey across 2024 through 2026 repeatedly identify people, process, governance, and organizational change management as the leading constraints on scaling enterprise AI, corroborating the tweet's emphasis on change management challenges and the need for dedicated AI leadership roles within business units.
Legacy system modernization is a documented enterprise priority
CIO surveys and digital-priorities research from Gartner, Forrester, IDC, and consulting firms confirm that fragmented on-premises and poorly modernized cloud systems remain a top obstacle, with enterprises investing heavily in API-based composable architectures and data platform work to enable agents to access unified data sources.
Interoperability standards are emerging as an industry imperative
Anthropic released the Model Context Protocol in November 2024, Google announced Agent-to-Agent protocol in April 2025, and the Linux Foundation formed the Agentic AI Foundation in December 2025 to steward these open standards — concrete evidence that multi-agent interoperability is a measurable and growing industry priority, just as the tweet describes.
Token and compute cost management is a real operational concern
Enterprise platforms including Microsoft Foundry and Azure OpenAI have introduced model routing, quota management, telemetry, and governance features specifically because large organizations must actively manage inference spend and balance trade-offs between cost, latency, and output quality — validating the tweet's "tokenmaxxing" observation about compute budget rationing.
Contradicting Evidence
Job displacement data contradicts the "not replacing jobs" narrative
While the tweet claims most companies are not talking about replacing jobs, the World Economic Forum's 2025 Future of Jobs Report found that approximately 41% of employers plan workforce downsizing in roles where AI can automate tasks — a substantial share that suggests many organizations are indeed contemplating headcount reductions, even if they publicly emphasize augmentation and new revenue use cases.
Augmentation framing may reflect corporate messaging rather than ground truth
McKinsey, Gartner, and major consulting firms often emphasize AI as a tool for augmentation and revenue generation in their public-facing research, but CFO surveys and internal planning documents frequently reveal cost-cutting and labor efficiency as primary motivations for AI investment, indicating a gap between public narrative and private strategy that nuances the tweet's optimistic framing.
Shadow AI and security risks complicate the agentic adoption story
Netskope's 2025-2026 Cloud and Threat reports document significant growth in shadow AI and agentic AI usage within enterprises, along with rising data-leak and security risks — a dimension the tweet does not address that could slow or complicate the agent deployment trajectory the author describes.
The "everyone is busier than ever" claim lacks systematic evidence
While the tweet asserts unanimous agreement that teams are working harder than ever due to AI, this observation appears anecdotal and is not substantiated by published workforce productivity studies or enterprise surveys; some research suggests AI tools are reducing time on specific tasks even as they create new categories of work, making the net productivity and workload picture more complex than presented.
Report an Issue
Found something wrong with this article? Let us know and we'll look into it.