The expansion of traditional enterprise-level software is often accompanied by large sales teams and lengthy implementation cycles. From initial contact to final deployment, it usually takes several months, involving multiple demonstrations, compliance reviews, and customization development. But AI employee Viktor breaks this norm.
Before delving into business data, it is necessary to clarify what Viktor actually is. This product was created by a development team with a background in DeepMind, and its core idea is to create a “Tier 3 AI Coworker,” rather than a simple Co-pilot. The Viktor team believes that current AI tools mostly remain in the “draft and wait for human completion” phase, while Viktor’s goal is “end-to-end execution and delivering results.”
In simple terms, Viktor is like a tireless digital employee. You don't need to teach it how to use various software, nor do you need to write complex commands. You only need to @ it in the chat box of Slack or Teams, telling it “help me check the sales data for the East China region last week and generate a report with charts,” and it will pull the data from the CRM system, generate charts in the spreadsheet tool, and send the final report back to the conversation window. Besides responding passively, it can also proactively work when triggered by certain times or events, such as automatically doing reconciliations late at night, or collecting data across six different tools to generate a board presentation.
According to official disclosures, this product achieved an annualized revenue of $20 million on the Slack platform without forming a sales team or having implementation projects, serving over 30,000 companies. Recently, Viktor officially integrated with Microsoft Teams, opening a free trial to an ecosystem of 320 million users. When AI employees abandon prompt engineering for “zero-threshold @mentions,” has the tipping point for enterprise automation arrived? This is not just a product feature update issue; it also concerns the underlying reconstruction of the business model for enterprise-level AI applications.
No Sales Team, $20 Million Income, PLG Model's Victory in Enterprise-Level AI
The enterprise-level SaaS industry has long adhered to a “sales-driven” model. To land major clients, companies need to build large sales teams, configure customer success managers, and go through prolonged POC (Proof of Concept) and implementation cycles. The customer acquisition costs of this model are extremely high and heavily reliant on maintaining interpersonal relationships. However, Viktor's performance on Slack showcased a completely different path.
Officially released data shows that Viktor achieved $20 million in annual revenue without forming a sales team, having implementation projects, or per-seat billing contracts, serving 30,000 companies. This pure PLG (Product-Led Growth) model has precedents in the traditional SaaS era, but is extremely rare in complex enterprise-level AI applications. AI products usually require extensive contextual configuration and scenario debugging, making it difficult to use out of the box. The reason Viktor can achieve self-propagation lies in its reduction of configuration thresholds to the minimum.
The traditional SaaS per-seat billing model often makes companies face the concern of “idle waste” during procurement. Buying 100 accounts may lead to only 20 people using them frequently, while the remaining 80 accounts become sunk costs. Viktor tends to charge based on credit or task consumption, which aligns more closely with the actual logic of AI executing tasks. Companies no longer pay for the “potential number of employees who might use AI” but instead for the “actual workload completed by AI.”
This billing method reduces the trial-and-error costs for enterprise procurement, allowing department-level managers and even frontline employees to start experimenting directly with credit cards or free credits, bypassing lengthy IT procurement approvals. The success of this business model verifies a judgment: the core barrier for enterprise-level AI products does not lie in the coverage capacity of sales channels but in whether the product itself can prove its value within a very short experience cycle.
Viktor's strategy of offering a $100 free credit without requiring a credit card is precisely aimed at maximizing the shortening of this “value validation” cycle. When employees discover that @ mentioning Viktor can complete reconciliation work that originally took several hours, the product's self-propagation naturally occurs. According to public reports, Viktor recently completed a $75 million Series A funding round led by DN Capital, which also reflects market recognition of its PLG model. However, it should be noted that the specific calculation criteria for the $20 million ARR have not been detailed by the official source, whether it is calculated based on credit consumption, action-based billing, or a mixed model remains unknown to outsiders. This lack of transparency in billing helps lower the trial threshold in the early stages, but could become a barrier to ROI assessment during large-scale enterprise procurement.
Flattening the Prompt Engineering Barrier: From “Draft and Wait” to “End-to-End Delivery”
The key to Viktor achieving zero configuration self-propagation lies in its dimensionality reduction of interaction paradigms. The effectiveness of traditional AI tools highly depends on the user’s ability to write prompts. An article on OmniTools titled “After Observing for Three Years, I Categorized Everyone’s AI Usage Level into Ten Grades” has detailed this phenomenon: from structured prompts to encapsulating agent skills, AI users’ levels are divided into multiple tiers, and prompt engineering becomes an invisible barrier.
This barrier is particularly deadly in real enterprise scenarios. Financial personnel, HR specialists, and operations managers do not have the time or obligation to learn how to engage in complex “prompt negotiations” with AI. If the effectiveness of AI usage depends on employees’ prompt crafting abilities, then AI can only ever be a productivity tool for a few geeks and not become a universal infrastructure for enterprises.
Viktor is positioned as a “Tier 3 AI Coworker” rather than simply a Co-pilot. The native Co-pilot logic operates on “draft and wait for human completion,” excelling at summarizing documents, drafting emails, but still requiring human intervention for the final step. For example, if you ask Co-pilot to write a follow-up email to a client, once it is completed, you need to copy it into the email client, manually fill in the recipient, and send it. Viktor’s logic is “end-to-end execution and delivering results.” Users only need to describe their goals in natural language; the agent will autonomously decide on the execution steps, invoking the necessary tools to complete the loop. Similarly, when following up with a client, Viktor can directly connect to the email system, automatically fill in client information, and send it, even scheduling the next reminder based on the client's response.
This mechanism directly flattens the tier barriers brought by prompt engineering. The effectiveness of AI usage no longer depends on employees’ prompt crafting skills, but rather on the clarity of business objectives. This interaction method pushes AI from being an “auxiliary tool” to an “executor,” allowing non-technical personnel to enjoy AI benefits without friction.
However, this does not mean Viktor is free from the risk of misunderstanding. When users describe goals using vague natural language, the AI’s runtime autonomous decision-making mechanism may produce execution paths that do not align with user expectations. For example, if a user says, “clean up the sales pipeline,” Viktor might automatically mark some long-neglected opportunities as “failed,” whereas this could require a more complex approval in the enterprise sales process. The zero threshold lowers the usage barrier, but also raises higher demands for the accuracy of describing business objectives.
Nightly Automatic Reconciliation and Cross-Tool PPT Generation: How AI Becomes the “Process Layer”
If @mentioning is a passive response to human commands, then Viktor’s automatic triggering mechanism demonstrates the proactivity of AI employees, which is also its core distinction from traditional chatbots. According to Viktor’s official disclosure, the product supports automatic triggering scenarios without manual @mentions, such as nightly reconciliation, marking errors in accounts, screening applicants, scheduling calls, generating board PPTs across six siloed tools, and running regular tasks at 5 a.m.
These scenarios reveal an important trend: AI is descending from the “dialogue layer” to the enterprise’s “process layer.” An article on OmniTools titled “With Daily Active Users Jumping to Three to Four Times the Second in the Industry, What Gap Did Tencent WorkBuddy Tear Open in Office Agents?” has explored how office agents serve non-developer groups. Whether it is Viktor or WorkBuddy, the core logic is to encapsulate fixed processes that originally required crossing multiple systems and numerous human steps into atomic tasks that AI can execute automatically.
Take financial reconciliation as an example; in traditional processes, finance personnel need to export payment data from Stripe, export accounting data from Xero, perform VLOOKUP comparisons in Excel to find discrepancies, and manually flag them. This process is tedious and time-consuming, typically taking finance personnel two hours. Viktor, through hosted authentication, connects with over 3,200 tools; when the system time reaches a set node at night, Viktor will automatically log into Stripe and Xero, pull the day's data, execute comparison logic, and send a report with flagged errors to the finance channel. The entire process requires no manual intervention, and according to officials, it only takes six minutes.
Moreover, consider the cross-tool generation of board member presentations. Executives require a brief that includes sales data, product progress, and market feedback. Traditionally, assistants would need to open CRM, project management tools, and customer service systems separately, copy data, create charts, and finally paste into the PPT. Viktor can automatically execute this entire series of actions at 5 a.m., directly outputting a complete PPT file in the chat window.
Supporting such automatic triggering capability is Viktor’s organizational-level memory and contextual awareness mechanism. According to third-party assessments, Viktor possess persistent memory. If finance personnel correct Viktor on UTM formats or reconciliation rules once, Viktor will remember it permanently and automatically apply that rule in all subsequent related tasks. It can even read channel conversation history and proactively explain past decision reasons.
This mechanism makes Viktor not just a task-executing tool but a “process layer” that embodies the best practices and business rules of the enterprise. It reduces the friction costs associated with manual reminders, handovers, and “emotional management.” When seasoned employees leave and new employees start, the rules and processes within Viktor’s memory remain, ensuring the continuity of business execution.
From Slack to Teams: How the PLG Model Navigates the Compliance Deep Water Zone
Viktor’s integration with Microsoft Teams marks a critical step in its commercialization process. Although Slack is known for its flexibility and developer-friendly nature, serving as a “testbed” for lean teams and front-line companies, Microsoft Teams has a more complete departmental structure, approval chains, and organizational charts, making it the home of “real large organizations.” Official data shows that Teams has 320 million users. Viktor’s entry into Teams marks the shift of AI employees from being “geek toys” to formally entering the “core procurement vision of enterprises.”
However, transitioning from Slack to Teams is not merely a platform migration but the beginning of the PLG model’s entrance into compliance deep waters. On Slack, users can complete app installation and authorization within seconds; such extremely low friction is the foundation for Viktor's viral propagation. However, in Teams, this quick installation is replaced by lengthy IT administrator approval queues, security audits (such as SOC 2 compliance requirements), and application governance policies.
IT departments in large enterprises remain highly vigilant towards any third-party application that has data read-write permissions. To achieve end-to-end task execution, Viktor must obtain read-write permissions for CRM, financial systems, and even code repositories. This high permission requires it to undergo the enterprise's procurement cycle. The “bottom-up” PLG propagation path validated by Viktor on Slack might be interrupted by the IT department’s “top-down” controls in Teams.
To cope with this challenge, Viktor similarly offers a $100 free credit trial on Teams without requiring a credit card. This is a typical “wedge” strategy aimed at allowing frontline employees to experience product value before the IT departments are aware, creating an internal voice that then pressures the IT department to undertake compliance approvals. However, the effectiveness of this strategy within the Teams ecosystem remains to be observed. After all, enterprise-level procurement decisions depend not only on product experiences but also on compliance risks and data asset security.
The Cost of Fully Automated Execution: Black Box Risks and Trust Games
The “zero-threshold” and “fully automated execution” vision outlined by Viktor undoubtedly hits the pain point of enterprise operational efficiency. However, during actual deployment, this model faces undeniable crises of trust and black box risks.
To achieve broad coverage and end-to-end delivery, Viktor sacrifices fine-grained control over every execution step. Traditional workflow automation tools (like n8n or Zapier) are complicated to configure, but every step's data flow and logical branches are visible, allowing operators to locate errors clearly. In contrast, Viktor’s runtime autonomous decision-making mechanism makes the execution process, to some extent, a “black box.” When the AI has “read-write permission” to CRM or financial systems, a model delusion or an erroneous interpretation of natural language instructions could lead to incorrect data being written into production systems, resulting in data pollution or even business disruptions.
Decision-makers in enterprise procurement are often most concerned about the risk of “misoperation.” If an AI employee can automatically update customer information in HubSpot or create invoices in Xero without strict per-user permissions and audit logs, a single erroneous execution could require significant human resources for data rollback and recovery. For instance, if Viktor mistakenly marks a batch of high-value opportunities as “failed” while automatically cleaning the sales pipeline, the sales team could lose important client leads, and such errors may not be discovered until days later.
To prevent these risks, companies often have to enable “audit prioritization default settings” for actual use. This means Viktor must wait for manual confirmation before executing critical write operations. While this compromise reduces risks, it also disrupts the vision of “fully automated and unattended” execution, reinstating steps of manual intervention. Finding a balance between “efficiency enhancement” and “misoperation disasters” is a question that all AI employee products must address.
Viktor’s automatic triggering mechanism also brings new management challenges. When AI can execute tasks automatically based on events or time, enterprises need to establish a new monitoring system to ensure that AI behaviors consistently align with business rules and compliance requirements. Strict permission management, detailed audit logs, and interpretable decision paths are prerequisites for the large-scale deployment of AI employees. If these issues remain unresolved, AI employees may forever stay on the margins of departmental scenarios, unable to truly enter the core business flows of enterprises.
From Slack to Teams, Viktor has validated the appeal of zero-threshold interactions in the enterprise-level market while exposing the compliance resistance of the PLG model in large organizations. For AI employees to truly become the infrastructure of enterprises, it requires not only smarter models and lower interaction thresholds but also a governance framework that can earn enterprise trust. Only when the balance between efficiency and security is gradually achieved will the tipping point for enterprise automation truly arrive.
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
