Whether you're looking for an angel investor, a growth advisor, or just want to connect — I'm always open to great ideas.
Get in TouchAI, startups & growth insights. No spam.
TL;DR: On March 1, 2026, Melbourne-based AI agency Enterprise Monkey publicly announced it was quitting ChatGPT over two specific triggers: OpenAI's Pentagon deal and the introduction of advertising inside ChatGPT. Their CEO, Aamir Qutub, is migrating all internal operations and the company's flagship Zee AI agent to Anthropic's Claude — and rewriting a published book to remove all ChatGPT references. Behind him: 700,000+ users who have already abandoned ChatGPT as part of the #QuitGPT movement, generating 36 million+ impressions on X. The original announcement is at GlobeNewswire.
Seven hundred thousand users do not abandon a dominant software platform over nothing.
When you hear that figure attached to a hashtag — #QuitGPT — the temptation is to dismiss it as social media noise. A viral moment, a Twitter pile-on, a news cycle that will fade by next week. That dismissal would be a mistake.
The #QuitGPT movement, which accelerated in late February and early March 2026, represents something that enterprise software markets rarely produce: a values-driven churn event. Not churn caused by a better competing product (though that argument is being made too). Not churn caused by a pricing increase or a service outage. Churn caused by a customer base that looked at what their AI vendor was doing and decided — consciously, publicly, sometimes loudly — that they no longer wanted to be associated with it.
700,000+ users abandoned ChatGPT. The hashtag generated 36 million+ impressions on X. And while most of those users are individuals, the movement has a meaningful enterprise signal embedded inside it: companies that built products, workflows, and client commitments on ChatGPT are now publicly migrating away.
Enterprise Monkey, a Melbourne-based AI agency, is among the most visible. Their announcement on March 1, 2026 was not a casual social post. It was a structured press release, published via GlobeNewswire, detailing exactly why they were leaving, what they were migrating to, and what that migration looked like in practice. CEO Aamir Qutub named two specific triggers. He articulated a technical rationale. He is rewriting a published book.
That level of commitment is not a PR stunt. It is a declaration about how this company intends to position itself in an AI market that is, for the first time, splitting along ethical fault lines.
This article examines what Enterprise Monkey did, why they did it, what it means for their business and their product, and — most importantly — what it signals for the broader enterprise AI market at a moment when OpenAI's dominance is no longer the only story worth telling.
Enterprise Monkey is a Melbourne-based AI agency. They are not a household name in the way that the companies they are reacting to are. They do not have the brand recognition of a Salesforce or an Accenture. What they have is something arguably more important for understanding where this story is going: they are a mid-market implementation shop. They are the kind of company that builds AI systems for other businesses, advises leadership teams on AI adoption, and stakes its own reputation on the tools it recommends.
That context matters enormously.
When a consumer quits ChatGPT, it affects their personal workflow. When a software agency quits ChatGPT, it affects every client they serve, every recommendation they make, and every product they ship. An agency migration is a multiplier event. One company switching vendor means dozens of downstream projects that will now be built on a different stack.
CEO Aamir Qutub has built his public profile around AI adoption for non-technical business leaders. He hosts a podcast called "The Dumb Monkey Show" and authored a book — "The CEO Who Mocked AI (Until It Made Him Millions)" — built around the narrative of a construction CEO transforming a business through AI. That book has been, until now, largely a ChatGPT story. Qutub's public persona was built in part on demonstrating what ChatGPT could do for enterprises that had been skeptical of AI.
That he is now rewriting that book to replace ChatGPT references with Claude tells you everything you need to know about how seriously he is taking this shift. This is not someone quietly swapping one API for another in their backend. This is someone revising the artifact that defines their public identity in the AI space.
For the enterprise market, that kind of signal is worth paying attention to.
Qutub was explicit in the announcement. There were two specific triggers:
1. OpenAI's Pentagon contract — announced in late February 2026, which gave the U.S. Department of Defense access to OpenAI's AI models on classified military networks.
2. The introduction of advertising inside ChatGPT — also in February 2026, which represented a shift in how OpenAI is monetizing its flagship consumer product.
These are not abstract philosophical objections. They are concrete product and business decisions that OpenAI made, and that Enterprise Monkey responded to with a concrete business decision of their own.
The combination of the two triggers is important. Either one alone might have been manageable. A defense contract is, on its own, a geopolitical position that different customers will evaluate differently depending on their own values and client base. Advertising in a productivity tool is, on its own, an annoyance that many users would tolerate in exchange for free or low-cost access.
Together, however, they paint a picture of an OpenAI that is prioritizing revenue and government relationships over the interests and trust of the enterprise customers who built their workflows on its platform. That picture, right or wrong, is one that a significant number of enterprise clients found unacceptable.
The OpenAI Pentagon deal has been covered extensively — including in our earlier piece on Caitlin Kalinowski's resignation — but the specific lens that matters here is the enterprise trust lens.
Enterprise clients, particularly those in regulated industries, professional services, and agencies that serve clients across multiple sectors, are acutely sensitive to the data and governance implications of the AI tools they use. When an AI vendor signs a classified military contract, it does not just change what that vendor is doing with government clients. It changes the nature of the vendor's relationship with data governance, with legal oversight, and with the transparency expectations that enterprise clients rely on.
The sequence of events around the deal made this worse. OpenAI announced the Pentagon agreement before the contractual guardrails were defined. The agreement was revised after the initial announcement triggered backlash — including the resignation of the company's own head of robotics, Caitlin Kalinowski, who cited the rushed process as a governance failure. Sam Altman publicly acknowledged the rollout appeared "opportunistic and sloppy."
For enterprise clients building AI-native products, that sequence is not reassuring. Enterprises run procurement processes specifically to evaluate vendor governance. When a vendor announces a major government contract, revises its terms under public pressure, and has a senior executive resign in protest — all within the span of two weeks — the procurement red flags multiply.
There is also a competitive dynamic worth noting. The Pentagon deal emerged hours after the Trump administration effectively banned Anthropic from defense contracts, labeling it a "supply-chain risk." Anthropic had reportedly demanded explicit contractual prohibitions on domestic surveillance and autonomous weapons as a condition of its own Pentagon negotiations. The Pentagon refused. Anthropic was blacklisted. OpenAI stepped in.
This sequence created an unusual market position: Anthropic, which had been excluded from defense contracts specifically because it insisted on stricter ethical guardrails, suddenly looked like the more trustworthy enterprise partner to a certain class of buyer. For Enterprise Monkey's Qutub, Anthropic's stance was explicitly part of the calculus: the announcement noted Anthropic's ethical position on refusing surveillance and autonomous weapons use as one of the reasons for switching to Claude.
The enterprise AI market is not monolithic. There are buyers who want their AI vendor deeply embedded in defense and government work. There are buyers who actively do not. The Pentagon deal sorted those buyer segments in a way that had not previously been visible — and for the segment that does not want that exposure, Anthropic is now the obvious alternative.
The advertising trigger is, in some ways, a more immediately practical concern for enterprise users than the defense contract.
When OpenAI introduced advertising into ChatGPT in February 2026, it changed the fundamental value proposition of the product. ChatGPT's enterprise appeal has always been, in part, the simplicity of the commercial relationship: you pay for API access or a subscription, OpenAI gives you an AI system, there is no third-party advertiser sitting between the user and the product.
Advertising changes that relationship. It introduces a new principal — the advertiser — whose interests are not necessarily aligned with the user's. In a consumer product, this is familiar and manageable. In an enterprise AI tool that is handling sensitive client data, driving autonomous business processes, and producing content that goes out under the client's brand, it is a different kind of problem.
Enterprise clients ask very specific questions about the AI tools in their stack: Where is my data going? Who can see it? What influences the outputs? When an AI product starts serving ads, a reasonable enterprise client starts asking whether advertiser relationships have any influence over the model's behavior, its recommendations, or the content it produces. Those questions may have clean answers. But the fact that they now have to be asked at all represents a trust erosion.
For an agency like Enterprise Monkey, which builds AI-powered content production, email automation, and CRM workflows for its clients, the advertising question has a direct business dimension. If their Zee agent is producing marketing content and communications on behalf of clients, and that agent is running on a model that serves advertising, the clients reasonably want to know what that means for the integrity of the outputs.
Claude does not serve advertising. Anthropic's commercial model is API access and enterprise subscriptions. That's a cleaner story to tell a skeptical client.
The #QuitGPT movement did not begin with Enterprise Monkey. By the time Qutub published his announcement on March 1, the movement was already generating significant volume. But the numbers are worth stating plainly because they challenge the assumption that this is a fringe reaction.
700,000+ users have abandoned ChatGPT as of early March 2026. This is not seven hundred thousand people deleting an app they barely used. The profile of #QuitGPT participants, based on the X/Twitter discourse, skews toward active users — developers, agency professionals, content creators, and small business owners who had integrated ChatGPT into their actual workflows.
36 million+ impressions on X means the conversation reached people well beyond the activists. In social media terms, 36 million impressions is not a niche movement. It is a mainstream news cycle.
For context: ChatGPT reportedly reached 300 million weekly active users as of late 2024. Seven hundred thousand departures represents a fraction of a percentage point of that base. Statistically, it is not existential for OpenAI's user count. But enterprise market dynamics do not work like consumer market dynamics.
In enterprise software, the decision-making unit is small — often a single CTO, CIO, or operations executive. Word travels through professional networks. An agency that publicly migrates away from a platform tells its clients, its prospects, and its industry peers something about where the safe bets are. The enterprise multiplier effect means that 700,000 individual departures may represent far more organizational influence than the raw number suggests.
The 36 million impressions number is arguably more important than the 700,000 departures. That is 36 million people who encountered the argument that ChatGPT is no longer the obvious default. For enterprise buyers who were not previously paying attention to the Pentagon deal or the advertising question, that exposure plants a seed that will surface in the next procurement conversation.
Enterprise Monkey's migration is not a marketing position. Qutub was specific about the scope: all internal AI operations, agents, and new product development have moved to Claude.
The technical rationale is explicit in the announcement. Claude was described as superior for autonomous agents because of:
These are not vague performance claims. They reflect specific technical characteristics of Claude that matter for the kind of work Enterprise Monkey does. Autonomous email agents, CRM automation, and content production pipelines all require a model that can maintain context across complex workflows, call external tools reliably, and produce outputs that do not require constant human review and correction.
Qutub is also explicit that the migration does not mean Enterprise Monkey will never recommend OpenAI to clients. The company maintains platform-independent recommendations and continues recommending OpenAI and Microsoft Copilot 365 where appropriate for client contexts. This is a sophisticated position — not a blanket rejection of OpenAI, but a deliberate restructuring of the company's own infrastructure and product around Claude, while maintaining the flexibility to recommend whatever is best for individual client situations.
That distinction is important. Enterprise Monkey is not claiming Claude is categorically superior for all use cases. They are saying that for the kind of autonomous, agentic, high-stakes enterprise work they build, Claude is now their default — and the reasons are both ethical and technical.
The Zee AI agent is the product-level manifestation of Enterprise Monkey's Claude migration. It is also the piece of the story that most directly speaks to the enterprise market implications.
Zee is a proprietary autonomous agent that manages email, CRM workflows, content production, and media outreach. It already runs entirely on Claude. For a company of Enterprise Monkey's profile — a boutique AI agency whose entire value proposition is the sophistication of the AI systems it builds — the architecture of Zee is not a back-office detail. It is the central proof point.
Building Zee on Claude-native infrastructure means the agent's behavior, its reliability, its integration patterns, and its improvement trajectory are all tied to Anthropic's platform. MCP integrations, which Qutub cited as a specific advantage, allow Zee to connect directly to external tools — email clients, CRM systems, content management platforms — using a standardized protocol that Anthropic has been promoting aggressively as an open standard for AI agent infrastructure.
The Claude-native positioning matters beyond just the current feature set. Anthropic has been investing heavily in agentic capabilities — the kind of multi-step, tool-using, context-maintaining behavior that autonomous enterprise agents require. By building Zee on Claude, Enterprise Monkey is betting that Anthropic's agentic development roadmap is the right one to align with.
That bet reflects a broader pattern in the enterprise AI market. The question is no longer just "which model produces better text?" The question is "which platform gives us the infrastructure to build reliable, auditable, maintainable enterprise agents?" The answer to that question is genuinely contested — and Enterprise Monkey has placed their chips on Anthropic.
The book detail is easy to skim past. It should not be.
Aamir Qutub's book, "The CEO Who Mocked AI (Until It Made Him Millions)," is a narrative about a construction company executive's AI transformation journey. The book's central technology, from the available description, has been ChatGPT. Qutub built his public reputation — the podcast, the agency, the thought leadership positioning — partly on demonstrating what ChatGPT could do for skeptical executives.
He is now rewriting that book. The new edition will replace ChatGPT references with Claude and broaden the focus to "ethical and sovereign AI alternatives."
Think about what that requires. Writing a book is a substantial investment of time and public credibility. A published book is a fixed artifact — something you stand behind, something clients and prospects read to understand your worldview. Revising a published book to change its central technology reference is not a small decision. It is a signal about the depth of conviction behind the migration.
An executive who was performatively jumping on the #QuitGPT bandwagon for marketing purposes would not rewrite a book. They would post a tweet, get some impressions, and move on. Rewriting a book is what someone does when they have genuinely changed their view and need the public record to reflect it.
The "ethical and sovereign AI alternatives" framing in the new edition is also worth noting. "Sovereign AI" is a phrase that has gained traction in enterprise conversations about AI governance — the idea that organizations should have meaningful control over the AI systems they depend on, without that control being subordinated to the vendor's government relationships, advertising partners, or regulatory constraints. By incorporating that framing into his book's premise, Qutub is not just repositioning for Claude. He is repositioning for a specific narrative about what enterprises should want from their AI vendors.
That narrative is commercially viable. The question is whether it reaches enough of the market to matter.
Let's be precise about what the Enterprise Monkey story is and is not.
It is not evidence that OpenAI is losing its enterprise market dominance. ChatGPT Enterprise and the OpenAI API remain the default starting point for the vast majority of enterprise AI deployments. The network effects, the integrations, the familiarity, and the sheer breadth of OpenAI's model portfolio give it an inertial advantage that a 700,000-user hashtag does not overcome.
What it is evidence of is the emergence of a distinct enterprise buyer segment — call it the "ethical AI" segment or the "sovereign AI" segment — that is now actively de-risking away from OpenAI in the wake of the Pentagon deal and the advertising expansion.
How large is that segment? Precise measurement is impossible, but the 36 million impressions number suggests it is large enough to constitute a real market. Enterprise software markets have multiple viable players with smaller shares than the dominant vendor. Anthropic does not need to topple OpenAI to build a substantial business on the buyers who, for values or governance reasons, prefer a vendor without defense contracts and without advertising in the product.
The enterprise AI consulting market is also a key lever here. Agencies like Enterprise Monkey do not just use AI tools — they recommend them, build on them, and stake their reputation on them. When agencies migrate, they carry client relationships with them. A mid-market agency that has moved its agentic infrastructure to Claude will naturally guide new clients toward Claude-based implementations, unless there is a specific reason to recommend otherwise.
This is how enterprise market shifts happen at the margins: not through mass simultaneous defection, but through the slow reorientation of the implementation layer — the agencies, consultants, and system integrators who make the actual buying decisions for their clients. Enterprise Monkey is one agency. But if the pattern holds across dozens or hundreds of similar shops, the cumulative effect on Anthropic's enterprise pipeline becomes material.
The timing also intersects with a structural shift in how enterprises think about AI vendor selection. The first wave of enterprise AI adoption was characterized by urgency: get something deployed, prove the ROI, figure out governance later. That phase is ending. Enterprises that deployed ChatGPT-based systems in 2023 and 2024 are now in the renewal and expansion cycle — the moment when procurement processes catch up with what was initially adopted on an experimental basis. At exactly that moment, OpenAI has handed skeptical procurement teams two new reasons to take a harder look at alternatives.
Anthropic's own positioning has evolved to exploit this opening. Its "Constitutional AI" approach, its explicit refusal of the Pentagon contract terms, and its aggressive push on enterprise features — including MCP and Claude's agentic capabilities — are all signals that Anthropic is actively competing for the buyer segment that #QuitGPT is surfacing. The fact that Enterprise Monkey's Qutub cited Anthropic's ethical stance as part of his rationale suggests that messaging is landing.
The addressable market for "enterprise AI that doesn't do defense contracts" is not a niche. It includes financial services firms with strict data governance requirements, healthcare organizations navigating HIPAA in an AI context, international clients in jurisdictions that have regulatory concerns about U.S. government access to AI infrastructure, and professional services firms whose client relationships depend on demonstrable neutrality. These are high-value, high-retention enterprise segments.
The Enterprise Monkey story is ultimately about what enterprises are buying when they buy AI — and how that calculus is shifting.
In the early adoption phase, enterprises bought capability. They bought the ability to do things with AI that they could not do before: generate content at scale, automate repetitive analysis, build internal chatbots, accelerate software development. Capability was the primary differentiator and ChatGPT, as the most capable and accessible system available, captured the dominant share.
That phase is not over. Capability still matters. But a second dimension is emerging in the buying conversation: alignment. Not alignment in the technical AI safety sense, but commercial and values alignment. Enterprises are starting to ask: does this vendor's strategic direction align with what we need from an AI partner over the next five years? Does their governance approach match ours? Are their revenue model and their client interests pulling in the same direction?
On those questions, the Pentagon deal and the advertising expansion create genuine friction. Not for every enterprise buyer — but for a meaningful segment, they shift the alignment calculus in Anthropic's direction.
Aamir Qutub's decision to rebuild Zee on Claude, rewrite his book, and announce the migration publicly is a bet that the alignment question will matter more in enterprise AI as the market matures. He is not the only person making that bet. He is simply one of the most visible.
The #QuitGPT movement's 700,000 departures are a lagging indicator of a leading indicator: the moment when enterprise buyers started asking whether OpenAI's strategic choices served their interests. That question will not stop being asked. The answer different enterprises give will determine whether the movement is a moment or the beginning of a meaningful market realignment.
For now, Enterprise Monkey has its answer. And it runs on Claude.
What is #QuitGPT? A social movement that accelerated in late February and early March 2026 following OpenAI's Pentagon defense contract announcement and the introduction of advertising in ChatGPT. Over 700,000 users have publicly abandoned the platform, generating 36 million+ impressions on X.
Who is Enterprise Monkey? A Melbourne-based AI agency led by CEO Aamir Qutub, which builds AI-powered business automation tools including the Zee autonomous agent. They announced their migration from ChatGPT to Anthropic's Claude on March 1, 2026.
What is the Zee AI agent? Enterprise Monkey's flagship autonomous AI agent that manages email, CRM workflows, content production, and media outreach. It now runs entirely on Claude.
Why did Enterprise Monkey switch to Claude specifically? Two reasons cited: ethical (Anthropic's refusal to include surveillance and autonomous weapons capabilities in the Pentagon deal) and technical (Claude's better structured reasoning, MCP integrations, native tool use, and fewer hallucinations for autonomous agent workflows).
Is ChatGPT losing its enterprise market dominance? Not in absolute terms — OpenAI remains the default for most enterprise deployments. However, a distinct "ethical/sovereign AI" buyer segment is emerging that is actively evaluating alternatives, and agencies like Enterprise Monkey serve as multipliers of that shift across their client base.
What does "sovereign AI" mean in this context? The concept that organizations should maintain meaningful control over the AI systems they depend on, without that control being subordinated to the vendor's government relationships, advertising partnerships, or regulatory exposure. It is an emerging framing in enterprise AI procurement conversations.
Is Enterprise Monkey still recommending ChatGPT to clients? Yes, in contexts where it remains the most appropriate tool. The company maintains platform-independent recommendations and continues recommending OpenAI and Microsoft Copilot 365 where appropriate. The migration applies to their own internal operations and product development.
Four major advertising agencies have integrated Anthropic's Claude into production workflows for SEO audits, creative briefs, and brand automation — signaling a major enterprise shift in AI adoption.
OpenAI has acquired Promptfoo, the open-source AI security testing platform used by 130,000 developers and 25% of Fortune 500 companies, to harden its agentic AI systems against prompt injection, jailbreaks, and adversarial attacks.
Anthropic's Claude Opus 4.6 discovered 22 Firefox vulnerabilities in 14 days, 14 high-severity. What this Mozilla partnership means for AI-driven security.