Fractile Raised $720 Million Because AI Inference Has a Memory Problem Nvidia's GPUs Cannot Solve
Amazon rebranded its enterprise AI workspace to AWS Quick this week, and DXC Technology's 115,000-employee deployment is the customer zero case that makes the competitive threat to Microsoft Copilot structurally real.
DXC Technology completed an enterprise-wide deployment of Amazon Quick — now rebranded AWS Quick — across its full workforce of 115,000 employees in 70 countries in February 2026. Within the first 20 days of deployment, DXC engineers built approximately 400 AI agents on the platform. The company simultaneously launched a DXC Amazon Quick Practice: a formal business unit dedicated to selling the same deployment playbook to enterprise clients. DXC's public framing is that it is "Customer Zero" — a service provider that runs its own tools in production before selling them to anyone else.
This week Amazon renamed the product AWS Quick, dropping the Amazon brand in favor of the AWS cloud brand. The name change is not cosmetic. AWS is Amazon's enterprise cloud business — the entity that holds FedRAMP authorization, HIPAA eligibility, SOC 2, and ISO 27001 certifications and serves the most compliance-sensitive workloads in the world. By moving the product under the AWS brand, Amazon is positioning Quick not as a consumer AI assistant that also works at enterprises, but as an enterprise-first AI workspace that happens to have a consumer-accessible entry tier. It is the clearest declaration yet that Amazon intends to compete with Microsoft for the enterprise AI workspace layer.
What AWS Quick Is and the Product Evolution That Got Here
AWS Quick has a complicated ancestry. It began in April 2024 as Amazon Q Business, a conversational AI assistant for enterprise users built on top of Amazon Bedrock. Q Business was Amazon's initial answer to Microsoft 365 Copilot: an AI assistant that could answer questions about company documents, internal knowledge bases, and connected data sources.
Amazon Q Business was competent but under-resourced relative to Microsoft's investment in Copilot. The product evolved into Amazon Quick Suite in October 2025, gaining agentic capabilities — the ability to take actions in connected systems rather than just answer questions — and a tighter desktop application experience. The April 2026 relaunch as Amazon Quick followed the What's Next with AWS event, introducing a redesigned interface, a Free tier requiring no AWS account, and native connectors for Google Workspace, Zoom, Airtable, Dropbox, and Microsoft Teams.
The November 2026 rebrand to AWS Quick is a go-to-market decision, not a product change. Amazon Quick had a brand confusion problem: "Amazon" connotes the consumer e-commerce business in most people's minds, while "AWS" connotes the enterprise cloud business that holds the compliance certifications enterprise IT procurement needs. Renaming the product AWS Quick moves it into the mental category where enterprise IT decision-makers already trust Amazon products — a category with a 13-year track record of building enterprise credibility through compliance depth and infrastructure reliability.
Pricing: Structurally Cheaper Than Copilot at Scale
AWS Quick's current pricing consists of three user-facing tiers plus a per-account infrastructure fee that is shared across the entire account:
| Tier | Per User / Month | Key Capabilities | Account Fee |
|---|---|---|---|
| Essentials | $3 | Q&A on connected data, 5 integrations | $250/month |
| Professional | $20 | Full agent capabilities, 50+ integrations | $250/month |
| Enterprise | $40 | Quick Automate, unlimited agents, dashboards | $250/month |
Overage pricing is $3 per agentic hour and $6 per research hour beyond included allocations. A 30-day trial waives all fees for up to 25 users, making it one of the more generous enterprise AI workspace trials in the market.
For comparison: Microsoft 365 Copilot is $30 per user per month as an add-on to M365 E3 or E5 subscriptions. M365 E3 itself costs $36 per user per month, putting the effective all-in cost at approximately $66 per user per month. ChatGPT Enterprise is priced around $60 per user per month on annual contract, though OpenAI does not publish this figure publicly.
At $20 per user per month for Professional — the tier most enterprise deployments would use — AWS Quick is structurally two to three times cheaper than both primary competitors. The $250 per-account infrastructure fee adds cost for small deployments but is negligible above roughly 300 users, where it represents less than $1 per user per month in overhead.
The enterprise AI spending data from Menlo Ventures found that enterprise AI total spend hit $177 billion in 2026, but only $12.6 billion flowed to AI-native software vendors — the rest went to incumbents expanding their AI surface. Microsoft's position in that dynamic is strong: it already owns the productivity suite enterprise employees use daily, and Copilot is delivered within that suite with no new purchasing decision required. AWS Quick's price advantage is real but insufficient on its own to displace that embeddedness. The question is whether $20 per user is low enough to seed adoption in organizations where Microsoft's ambient integration is not a lock-in — Google Workspace shops, mixed-stack enterprises, and departments that want agentic automation without additional Power Automate overhead.
The DXC Customer Zero Deployment: What 115,000 Users Actually Proves
The DXC deployment is the most credible enterprise-scale proof point AWS Quick has. DXC Technology is a $13 billion IT services company with operations in 70 countries and clients across financial services, healthcare, and government. Rolling out any enterprise software to 115,000 employees across that geography is a non-trivial deployment — it requires user provisioning, identity federation, compliance certification across multiple national regulatory environments, endpoint configuration, help desk training, and change management at a scale most vendors never see in a single customer.
DXC's public statements on the deployment focus on the agent-building rate: approximately 400 agents in 20 days. That figure tells a specific story about the deployment's character. It was not a pilot — it was a production deployment where engineers immediately started building automation rather than evaluating whether the tool was viable. The DXC Amazon Quick Practice that launched alongside the deployment is evidence of commercial intent: DXC was not just deploying for its own use, it was building a practice to sell the same deployment approach to external clients. As of the launch announcement, more than 40,000 DXC engineers were already actively using an AI assistant connected to the Quick platform.
The specific customers AWS has named in its enterprise marketing — 3M, BMW, Mondelez, New York Life, Southwest Airlines — are a deliberately cross-vertical reference set. 3M and BMW are industrial manufacturers with complex internal knowledge management needs. Mondelez is a consumer packaged goods company with a large distributed workforce. New York Life is a heavily regulated financial services entity where HIPAA and SOC 2 compliance are formal procurement prerequisites. Southwest Airlines has specific air transport regulatory requirements. This reference portfolio communicates that Quick has been validated against the compliance requirements of multiple regulated industries, not just technology companies willing to accept early-product risk.
Quick Automate and the Agentic Enterprise
The most strategically important feature in AWS Quick's current product is Quick Automate, available on the Enterprise tier at $40 per user per month. Quick Automate extends Quick from a conversational AI assistant into an agent that can initiate actions in connected enterprise systems — creating tickets in Jira, updating records in Salesforce, triggering workflows in ServiceNow, or sending notifications in Slack — without requiring human intervention at each step.
This capability is the key differentiator in the enterprise AI workspace competition, and it is the feature that separates AWS Quick's Enterprise tier from Microsoft 365 Copilot's default capability. Copilot's declarative agents, available as part of Copilot Studio, provide similar functionality but require Microsoft Power Automate licenses and configuration that adds friction for most enterprise deployments. Quick Automate is bundled into the Enterprise tier without additional configuration requirements — a packaging decision that lowers the barrier for enterprise IT teams evaluating agentic workflow automation.
Oracle's AI Agent Studio, launched inside Oracle Fusion Cloud ERP, addresses the same workflow automation problem in a narrow vertical — Oracle's own enterprise applications. Quick Automate's scope is horizontal: it can connect to any application with an available integration. The horizontal scope is both an advantage — broader applicability — and a challenge: more complex to configure well for any specific workflow, and requiring more enterprise governance attention to manage the permission scope it creates.
The enterprise security concerns around agentic AI that have driven investment in AI firewall companies are directly relevant to Quick Automate. When Quick Automate has permission to take actions in Jira, Salesforce, and Slack, it creates the same agent permission surface that AIR Security's firewall is designed to control. Enterprise buyers deploying Quick Automate at scale should evaluate whether their existing security tooling covers the agent permission scope the tool creates — this is not a Quick-specific concern, but Quick Automate at $40 per user will accelerate agentic deployment faster than many enterprise security policies have prepared for.
The Compliance Moat: Why the AWS Rebrand Is Strategic
The rebrand to AWS Quick is about more than name recognition. The AWS brand carries a specific compliance portfolio that Amazon Quick, as a standalone product identity, did not unambiguously communicate. That compliance portfolio is the actual competitive differentiator in the enterprise markets Quick needs to win.
AWS Quick is HIPAA eligible, meaning Amazon will execute Business Associate Agreements allowing healthcare organizations to use it for workloads involving protected health information. It holds FedRAMP authorization, which is required for US federal government deployments and is a credibility signal for state, local, and regulated private-sector buyers. It carries SOC 2 Type II and ISO 27001 certifications, which satisfy the data security controls that enterprise IT procurement and legal teams require before approving a new SaaS tool.
Microsoft's enterprise AI governance through Execution Containers addresses security at the agent execution layer — a different problem from data compliance. Copilot's compliance story is built on M365's existing compliance portfolio, which is comprehensive for most enterprise needs. AWS Quick's compliance is built on the AWS cloud infrastructure that already serves US intelligence community cloud, national healthcare systems, and global financial institutions. By rebranding to AWS Quick, Amazon makes that provenance explicit in the product name — shortening the conversation with regulated-industry procurement teams who already have a compliance relationship with AWS.
How Amazon Plans to Win Against Microsoft's Productivity Lock-In
Microsoft's advantage in the enterprise AI workspace competition is not primarily about Copilot's quality relative to Quick. It is about the switching cost embedded in Microsoft 365. Enterprise employees spend most of their working hours in Teams, Outlook, Word, Excel, and SharePoint. Copilot is integrated into all of these applications. It is there when an employee opens Word; it offers to summarize an email when they open Outlook. The integration is ambient — it requires no deliberate choice to use it.
AWS Quick does not have this ambient integration. It is a separate application that employees must choose to open. Its integration with Microsoft Office applications is limited by the competitive relationship between Amazon and Microsoft. The Microsoft Teams connector in Quick is real, but Quick cannot appear inside Teams the way Copilot does.
Amazon's distribution strategy for Quick against this structural disadvantage has four components:
1. Price the entry barrier low enough to seed usage organically. The $3 Essentials tier and the 30-day free trial for 25 users are designed to get Quick into enterprise environments without requiring a formal procurement decision. When employees or teams adopt Quick on the Essentials tier, they create a user base that IT can formalize into Professional or Enterprise contracts — the same bottoms-up growth mechanic that drove Slack's enterprise penetration before Microsoft Teams emerged as the alternative.
2. Build the compliance story into the brand. The rebrand to AWS Quick, combined with pre-validated FedRAMP, HIPAA, and SOC 2 certifications, is designed to shorten the enterprise IT approval cycle. The harder it is for security and compliance teams to say no to Quick, the lower the internal friction for procurement — particularly in regulated industries where Microsoft's consumer brand association creates its own due diligence burden.
3. Use DXC as a channel. The DXC Amazon Quick Practice creates a channel of enterprise IT consultants who have deployed Quick in production, understand its limitations, and can guide enterprise clients through procurement and configuration. DXC's access to Fortune 500 and Global 2000 clients is a sales motion Amazon does not need to build internally. The Customer Zero strategy pays for itself when the customer becomes a reseller.
4. Win the agentic frame before Microsoft closes the gap. Quick Automate's enterprise-tier positioning signals that Amazon's thesis is that the enterprise AI workspace market will be won at the agentic automation layer, not the conversational assistant layer. If that thesis is correct — and the Menlo Ventures enterprise AI report supports it, with agent-related spending growing faster than assistant spending — Quick Automate's availability at $40 per user is a better price point than Copilot Studio's comparable functionality, which requires additional Power Automate licensing.
What Enterprise IT Teams Should Do Before Q1 2027
The enterprise AI workspace market in late 2026 is a three-way race between Microsoft 365 Copilot, ChatGPT Enterprise, and AWS Quick, with Gemini Enterprise in a secondary position. No single platform is dominant across all use cases, and most large enterprises will end up with more than one deployed — different departments adopting different tools based on existing stack relationships.
For enterprise IT teams evaluating AWS Quick specifically:
1. Run the 30-day free trial with a real 25-person team before any procurement commitment. The free trial is generous enough to run a real pilot. Use it on a team that already has significant internal documentation and a clear workflow automation need — not a team evaluating the product, but a team that has work to do with it.
2. Evaluate the data connector coverage against your actual stack. Quick's value is proportional to the data it can access. If your enterprise runs on Microsoft 365 and SharePoint, Quick's access to those data sources is limited relative to Copilot's native integration. If your enterprise runs on Google Workspace or a mixed environment, Quick's native Google Workspace connector is a material advantage.
3. Involve your security team early on Quick Automate. The agent execution permissions Quick Automate requires — write access to Jira, Salesforce, Slack, and other connected systems — create a security scope that most organizations do not currently manage in their AI governance policies. Before deploying Quick Automate at scale, map the permission scope and add it to your AI governance framework.
4. Model the total cost at the tier your use case actually requires. The $3 Essentials tier is not adequate for most enterprise agentic use cases. At $20 Professional or $40 Enterprise plus the $250 account fee, model the actual per-user cost for your expected deployment scale before comparing to Copilot's fully loaded cost.
5. Treat the DXC practice as a resource if you use DXC for managed services. DXC's enterprise Quick practice has deployment experience that most organizations do not have internally. If your organization already has a DXC relationship, the Practice is an accessible way to accelerate deployment and reduce configuration risk without building internal expertise from zero.
Takeaway: The rebrand to AWS Quick is Amazon's clearest declaration that it is competing for the enterprise AI workspace — not as a niche alternative to Microsoft Copilot, but as the enterprise cloud provider's answer to the ambient AI layer that Microsoft has built into Microsoft 365. DXC's 115,000-employee deployment proves that Quick can operate at full enterprise scale across multiple geographies and compliance regimes. At $20 per user per month, Quick is structurally cheaper than every comparable platform. The price advantage alone will not dislodge Microsoft's productivity lock-in, but it lowers the barrier for mixed-stack enterprises, Google Workspace shops, and departments that want agentic automation without additional Power Automate overhead. For enterprise IT teams, the decision is not Quick versus Copilot — it is whether the use cases you're solving for require Microsoft's ambient Office integration or whether Quick's compliance depth and pricing economics make it the better answer. That evaluation should start this quarter, not after Q1 budget approvals.
Frequently Asked Questions
What is AWS Quick and how is it different from Amazon Q Business?
AWS Quick is Amazon's enterprise AI workspace, rebranded from Amazon Quick in November 2026. It is the fourth iteration of the same product: Amazon Q Business launched in April 2024 as a conversational AI assistant for enterprise users, evolved into Amazon Quick Suite in October 2025 with agentic capabilities, relaunched as Amazon Quick in April 2026 with a redesigned interface and broader connector ecosystem, and was renamed AWS Quick in November 2026 to signal enterprise cloud positioning under the AWS brand. The functional difference between Amazon Q Business and AWS Quick is significant: Q Business was primarily a conversational assistant that answered questions about connected data sources, while Quick in its current form includes Quick Automate, an agentic workflow automation capability that can take actions in connected enterprise systems — creating tickets, updating records, triggering workflows — without requiring human approval at each step. The compliance portfolio — HIPAA eligibility, FedRAMP, SOC 2, ISO 27001 — has been consistent since Q Business but is now front-and-center in the AWS Quick brand identity.
How much does AWS Quick cost compared to Microsoft 365 Copilot?
AWS Quick has three tiers: Essentials at $3 per user per month, Professional at $20, and Enterprise at $40, plus a $250 per-account infrastructure fee each month. Overage pricing runs $3 per agentic hour and $6 per research hour beyond included allocations. A 30-day trial waives all fees for up to 25 users. Microsoft 365 Copilot is priced at $30 per user per month as an add-on to M365 E3 or E5 subscriptions, which themselves cost $36 and $57 per user per month. The effective all-in cost of Copilot for an organization running M365 E3 is approximately $66 per user per month. At $20 per user for the Professional tier — the tier most enterprise deployments would use — AWS Quick is two to three times cheaper than the comparable Microsoft offering. The $250 per-account infrastructure fee adds cost for small deployments but is negligible at anything above roughly 300 users.
How did DXC Technology deploy AWS Quick to 115,000 employees?
DXC Technology completed an enterprise-wide deployment of Amazon Quick — now AWS Quick — across its full workforce of 115,000 employees in 70 countries in February 2026. The deployment covered user provisioning, identity federation, compliance certification across multiple national regulatory environments, endpoint configuration, help desk training, and enterprise change management. DXC's public position is that it is 'Customer Zero' — a service provider running its own tools before selling them to enterprise clients. Within the first 20 days of deployment, DXC engineers built approximately 400 AI agents on the platform. DXC simultaneously launched a formal DXC Amazon Quick Practice, a business unit dedicated to selling the same deployment playbook to external enterprise clients. The Practice provides enterprise buyers with access to consultants who have completed a full 115,000-employee production deployment — a credibility advantage that shortens enterprise sales cycles significantly, particularly in procurement processes where reference customer requirements are formal prerequisites.
What is Quick Automate and what can it do in enterprise workflows?
Quick Automate is the agentic workflow automation capability bundled into AWS Quick's Enterprise tier at $40 per user per month. It enables Quick to take actions in connected enterprise systems — creating Jira tickets, updating Salesforce records, triggering ServiceNow workflows, sending Slack notifications — without requiring human approval at each step. This distinguishes Quick from earlier versions that were limited to answering questions about connected data sources. The key practical use cases are process automation where a user describes what they want accomplished in natural language and Quick executes the multi-step workflow, exception handling where Quick monitors a data source and initiates a workflow when a defined condition occurs, and report generation where Quick pulls data from multiple connected sources, synthesizes it, and delivers the output to a specified destination. Quick Automate connects to 50+ integrations in the Professional tier and more in Enterprise. The comparable Microsoft capability — Power Automate with Copilot Studio agents — requires additional licensing on top of the base Copilot subscription.
Is AWS Quick HIPAA compliant and enterprise-secure enough for regulated industries?
AWS Quick is HIPAA eligible, meaning Amazon will execute Business Associate Agreements allowing healthcare organizations to use it for workloads involving protected health information. It also holds FedRAMP authorization, SOC 2 Type II certification, and ISO 27001 certification. These are prerequisites for US federal government deployments, heavily regulated financial services procurement, and healthcare system adoption. The compliance portfolio is built on AWS's underlying cloud infrastructure, which serves US intelligence community cloud workloads, major financial institutions, and national health systems. For enterprise IT procurement teams evaluating Quick against compliance checklists, the certifications are pre-validated rather than requiring a customer-specific audit — which shortens procurement timelines significantly. The rebrand from Amazon Quick to AWS Quick in November 2026 is partly about making this compliance provenance explicit: AWS as a brand identity carries direct association with the compliance portfolio that Amazon as a consumer brand does not.
What are the main weaknesses of AWS Quick compared to Microsoft Copilot?
AWS Quick's primary structural disadvantage is ambient integration. Microsoft 365 Copilot is integrated into Teams, Outlook, Word, Excel, and SharePoint — the applications enterprise employees use most of their working day. Copilot surfaces inside these applications without requiring any deliberate choice; it is there when an employee opens Word or receives an email in Outlook. AWS Quick is a separate application that employees must choose to open, and its access to Microsoft Office application data is limited by the competitive relationship between Amazon and Microsoft. Quick's Teams connector is functional but Quick cannot appear inside Teams the way Copilot does. A second weakness is the agentic workflow maturity: Quick Automate is newer than Copilot Studio and has fewer enterprise deployment case studies. Organizations with complex multi-step automation requirements may find Copilot Studio's larger ecosystem of pre-built connectors and existing Power Automate flows more immediately usable. The price advantage Quick holds is real but insufficient on its own to overcome these embedded-stack advantages for organizations that have standardized on Microsoft 365.